Upload
others
View
3
Download
0
Embed Size (px)
Citation preview
onlineforlifeorg
6 Steps to Launching Effective Tests on Your Website
Featuring Our Friends From
onlineforlifeorg
A Few Quick Thingsbull Thank you We are delighted that
have made time to join us todaybull A video recording will be made
availablebull Also links to all of the experiments
and additional resources will be sentbull We want your questions
Nonprofit Innovation amp Optimization Summit
What A conference to advance the future of fundraising
When September 28-29
Where Fort Worth Texas
Tickets NIOSummitcom
Top Thought Leaders and Practitioners
Every Attendee will get an Extra $100 offDiscount Code SEPT2016wwwNIOSummitcom
Nonprofit Innovation and Optimization Summit
onlineforlifeorg
Todayrsquos Speakers
Tim KachuriakChief Innovation amp Optimization Officer NextAfter
Jason GSellLaunch Manager Optimizely
Amish S PatelLaunch Manager Optimizely
onlineforlifeorg
6 Steps to Launching Effective Tests on Your Website
Featuring Our Friends From
onlineforlifeorg
Before We Talk About Howto Test Letrsquos Start With
Where to Test What to Test
Three Key Areas to Focus Your Testing
Email Signup Pages Landing Pages Donation Pages
Three Key Areas to Focus Your Testing
Email Signup Pages Landing Pages Donation Pages
What to Test Value PropositionCopyAdding CopyRemoving CopyDifferent Copy
Number of Form FieldsButton Text
Optimizing Email Signup Pages
Email Signup Pages
onlineforlifeorg
How communicating value of an email newsletter impacts email acquisitionExperiment 1621
Communicating Value
Backgroundbull Partner Alliance Defending
Freedombull Tested email acquisition on
the home pagebull Home page receives
~12000 visitors each month with less than 1 giving their email
Control
Communicating ValueTreatment
Treatment Changesbull Added a call to action to the
headlinebull Included value proposition
language identifying the benefits of giving an email
bull Button communicated value
Communicating ValueControl Treatment
441 increase in emails acquired
Communicating Value
Version Conv Rate Relative Diff Stat Confidence
C Simple form 09T1 Value proposition form 13 441 999Key Learning
Every interaction with a visitor is a potential value exchange By communicating the benefits and value to the visitor we increasedconversion by 441
441 increase in emails acquired
Sometimes You Have to Eliminate Copy
Control Treatment
Experiment 2313
262 increase in emails acquired
Itrsquos Not Just Copy But the Right Copy
Control Treatment
295 decrease in emails acquiredExperiment 2606
onlineforlifeorg
How the amount of required information impacts conversionExperiment 289
Impact of Required Form Fields
Backgroundbull Client Hillsdale Collegebull Name acquisition offer for their
Imprimis publicationbull The form historically required both
email and home address so it could be send through mail and online
Control
Impact of Required Form Fields
Tested Elementsbull Reduced the required form
fields by removing address information
bull Removed image of Imprimisissues
Treatment
Treatment
Impact of Required Form Fields
1360 increasein conversion rate
Control
Impact of Required Form Fields
1360 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Full Address 322T1 Email Only 760 1360 1000
Key Learning
By removing the friction associated with the home address fields Hillsdale was able to more than double the number of emails acquired
Reducing Form Field Test
Control Treatment
483 increasein conversion rate
Experiment 4699
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
onlineforlifeorg
A Few Quick Thingsbull Thank you We are delighted that
have made time to join us todaybull A video recording will be made
availablebull Also links to all of the experiments
and additional resources will be sentbull We want your questions
Nonprofit Innovation amp Optimization Summit
What A conference to advance the future of fundraising
When September 28-29
Where Fort Worth Texas
Tickets NIOSummitcom
Top Thought Leaders and Practitioners
Every Attendee will get an Extra $100 offDiscount Code SEPT2016wwwNIOSummitcom
Nonprofit Innovation and Optimization Summit
onlineforlifeorg
Todayrsquos Speakers
Tim KachuriakChief Innovation amp Optimization Officer NextAfter
Jason GSellLaunch Manager Optimizely
Amish S PatelLaunch Manager Optimizely
onlineforlifeorg
6 Steps to Launching Effective Tests on Your Website
Featuring Our Friends From
onlineforlifeorg
Before We Talk About Howto Test Letrsquos Start With
Where to Test What to Test
Three Key Areas to Focus Your Testing
Email Signup Pages Landing Pages Donation Pages
Three Key Areas to Focus Your Testing
Email Signup Pages Landing Pages Donation Pages
What to Test Value PropositionCopyAdding CopyRemoving CopyDifferent Copy
Number of Form FieldsButton Text
Optimizing Email Signup Pages
Email Signup Pages
onlineforlifeorg
How communicating value of an email newsletter impacts email acquisitionExperiment 1621
Communicating Value
Backgroundbull Partner Alliance Defending
Freedombull Tested email acquisition on
the home pagebull Home page receives
~12000 visitors each month with less than 1 giving their email
Control
Communicating ValueTreatment
Treatment Changesbull Added a call to action to the
headlinebull Included value proposition
language identifying the benefits of giving an email
bull Button communicated value
Communicating ValueControl Treatment
441 increase in emails acquired
Communicating Value
Version Conv Rate Relative Diff Stat Confidence
C Simple form 09T1 Value proposition form 13 441 999Key Learning
Every interaction with a visitor is a potential value exchange By communicating the benefits and value to the visitor we increasedconversion by 441
441 increase in emails acquired
Sometimes You Have to Eliminate Copy
Control Treatment
Experiment 2313
262 increase in emails acquired
Itrsquos Not Just Copy But the Right Copy
Control Treatment
295 decrease in emails acquiredExperiment 2606
onlineforlifeorg
How the amount of required information impacts conversionExperiment 289
Impact of Required Form Fields
Backgroundbull Client Hillsdale Collegebull Name acquisition offer for their
Imprimis publicationbull The form historically required both
email and home address so it could be send through mail and online
Control
Impact of Required Form Fields
Tested Elementsbull Reduced the required form
fields by removing address information
bull Removed image of Imprimisissues
Treatment
Treatment
Impact of Required Form Fields
1360 increasein conversion rate
Control
Impact of Required Form Fields
1360 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Full Address 322T1 Email Only 760 1360 1000
Key Learning
By removing the friction associated with the home address fields Hillsdale was able to more than double the number of emails acquired
Reducing Form Field Test
Control Treatment
483 increasein conversion rate
Experiment 4699
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Nonprofit Innovation amp Optimization Summit
What A conference to advance the future of fundraising
When September 28-29
Where Fort Worth Texas
Tickets NIOSummitcom
Top Thought Leaders and Practitioners
Every Attendee will get an Extra $100 offDiscount Code SEPT2016wwwNIOSummitcom
Nonprofit Innovation and Optimization Summit
onlineforlifeorg
Todayrsquos Speakers
Tim KachuriakChief Innovation amp Optimization Officer NextAfter
Jason GSellLaunch Manager Optimizely
Amish S PatelLaunch Manager Optimizely
onlineforlifeorg
6 Steps to Launching Effective Tests on Your Website
Featuring Our Friends From
onlineforlifeorg
Before We Talk About Howto Test Letrsquos Start With
Where to Test What to Test
Three Key Areas to Focus Your Testing
Email Signup Pages Landing Pages Donation Pages
Three Key Areas to Focus Your Testing
Email Signup Pages Landing Pages Donation Pages
What to Test Value PropositionCopyAdding CopyRemoving CopyDifferent Copy
Number of Form FieldsButton Text
Optimizing Email Signup Pages
Email Signup Pages
onlineforlifeorg
How communicating value of an email newsletter impacts email acquisitionExperiment 1621
Communicating Value
Backgroundbull Partner Alliance Defending
Freedombull Tested email acquisition on
the home pagebull Home page receives
~12000 visitors each month with less than 1 giving their email
Control
Communicating ValueTreatment
Treatment Changesbull Added a call to action to the
headlinebull Included value proposition
language identifying the benefits of giving an email
bull Button communicated value
Communicating ValueControl Treatment
441 increase in emails acquired
Communicating Value
Version Conv Rate Relative Diff Stat Confidence
C Simple form 09T1 Value proposition form 13 441 999Key Learning
Every interaction with a visitor is a potential value exchange By communicating the benefits and value to the visitor we increasedconversion by 441
441 increase in emails acquired
Sometimes You Have to Eliminate Copy
Control Treatment
Experiment 2313
262 increase in emails acquired
Itrsquos Not Just Copy But the Right Copy
Control Treatment
295 decrease in emails acquiredExperiment 2606
onlineforlifeorg
How the amount of required information impacts conversionExperiment 289
Impact of Required Form Fields
Backgroundbull Client Hillsdale Collegebull Name acquisition offer for their
Imprimis publicationbull The form historically required both
email and home address so it could be send through mail and online
Control
Impact of Required Form Fields
Tested Elementsbull Reduced the required form
fields by removing address information
bull Removed image of Imprimisissues
Treatment
Treatment
Impact of Required Form Fields
1360 increasein conversion rate
Control
Impact of Required Form Fields
1360 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Full Address 322T1 Email Only 760 1360 1000
Key Learning
By removing the friction associated with the home address fields Hillsdale was able to more than double the number of emails acquired
Reducing Form Field Test
Control Treatment
483 increasein conversion rate
Experiment 4699
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Top Thought Leaders and Practitioners
Every Attendee will get an Extra $100 offDiscount Code SEPT2016wwwNIOSummitcom
Nonprofit Innovation and Optimization Summit
onlineforlifeorg
Todayrsquos Speakers
Tim KachuriakChief Innovation amp Optimization Officer NextAfter
Jason GSellLaunch Manager Optimizely
Amish S PatelLaunch Manager Optimizely
onlineforlifeorg
6 Steps to Launching Effective Tests on Your Website
Featuring Our Friends From
onlineforlifeorg
Before We Talk About Howto Test Letrsquos Start With
Where to Test What to Test
Three Key Areas to Focus Your Testing
Email Signup Pages Landing Pages Donation Pages
Three Key Areas to Focus Your Testing
Email Signup Pages Landing Pages Donation Pages
What to Test Value PropositionCopyAdding CopyRemoving CopyDifferent Copy
Number of Form FieldsButton Text
Optimizing Email Signup Pages
Email Signup Pages
onlineforlifeorg
How communicating value of an email newsletter impacts email acquisitionExperiment 1621
Communicating Value
Backgroundbull Partner Alliance Defending
Freedombull Tested email acquisition on
the home pagebull Home page receives
~12000 visitors each month with less than 1 giving their email
Control
Communicating ValueTreatment
Treatment Changesbull Added a call to action to the
headlinebull Included value proposition
language identifying the benefits of giving an email
bull Button communicated value
Communicating ValueControl Treatment
441 increase in emails acquired
Communicating Value
Version Conv Rate Relative Diff Stat Confidence
C Simple form 09T1 Value proposition form 13 441 999Key Learning
Every interaction with a visitor is a potential value exchange By communicating the benefits and value to the visitor we increasedconversion by 441
441 increase in emails acquired
Sometimes You Have to Eliminate Copy
Control Treatment
Experiment 2313
262 increase in emails acquired
Itrsquos Not Just Copy But the Right Copy
Control Treatment
295 decrease in emails acquiredExperiment 2606
onlineforlifeorg
How the amount of required information impacts conversionExperiment 289
Impact of Required Form Fields
Backgroundbull Client Hillsdale Collegebull Name acquisition offer for their
Imprimis publicationbull The form historically required both
email and home address so it could be send through mail and online
Control
Impact of Required Form Fields
Tested Elementsbull Reduced the required form
fields by removing address information
bull Removed image of Imprimisissues
Treatment
Treatment
Impact of Required Form Fields
1360 increasein conversion rate
Control
Impact of Required Form Fields
1360 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Full Address 322T1 Email Only 760 1360 1000
Key Learning
By removing the friction associated with the home address fields Hillsdale was able to more than double the number of emails acquired
Reducing Form Field Test
Control Treatment
483 increasein conversion rate
Experiment 4699
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Every Attendee will get an Extra $100 offDiscount Code SEPT2016wwwNIOSummitcom
Nonprofit Innovation and Optimization Summit
onlineforlifeorg
Todayrsquos Speakers
Tim KachuriakChief Innovation amp Optimization Officer NextAfter
Jason GSellLaunch Manager Optimizely
Amish S PatelLaunch Manager Optimizely
onlineforlifeorg
6 Steps to Launching Effective Tests on Your Website
Featuring Our Friends From
onlineforlifeorg
Before We Talk About Howto Test Letrsquos Start With
Where to Test What to Test
Three Key Areas to Focus Your Testing
Email Signup Pages Landing Pages Donation Pages
Three Key Areas to Focus Your Testing
Email Signup Pages Landing Pages Donation Pages
What to Test Value PropositionCopyAdding CopyRemoving CopyDifferent Copy
Number of Form FieldsButton Text
Optimizing Email Signup Pages
Email Signup Pages
onlineforlifeorg
How communicating value of an email newsletter impacts email acquisitionExperiment 1621
Communicating Value
Backgroundbull Partner Alliance Defending
Freedombull Tested email acquisition on
the home pagebull Home page receives
~12000 visitors each month with less than 1 giving their email
Control
Communicating ValueTreatment
Treatment Changesbull Added a call to action to the
headlinebull Included value proposition
language identifying the benefits of giving an email
bull Button communicated value
Communicating ValueControl Treatment
441 increase in emails acquired
Communicating Value
Version Conv Rate Relative Diff Stat Confidence
C Simple form 09T1 Value proposition form 13 441 999Key Learning
Every interaction with a visitor is a potential value exchange By communicating the benefits and value to the visitor we increasedconversion by 441
441 increase in emails acquired
Sometimes You Have to Eliminate Copy
Control Treatment
Experiment 2313
262 increase in emails acquired
Itrsquos Not Just Copy But the Right Copy
Control Treatment
295 decrease in emails acquiredExperiment 2606
onlineforlifeorg
How the amount of required information impacts conversionExperiment 289
Impact of Required Form Fields
Backgroundbull Client Hillsdale Collegebull Name acquisition offer for their
Imprimis publicationbull The form historically required both
email and home address so it could be send through mail and online
Control
Impact of Required Form Fields
Tested Elementsbull Reduced the required form
fields by removing address information
bull Removed image of Imprimisissues
Treatment
Treatment
Impact of Required Form Fields
1360 increasein conversion rate
Control
Impact of Required Form Fields
1360 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Full Address 322T1 Email Only 760 1360 1000
Key Learning
By removing the friction associated with the home address fields Hillsdale was able to more than double the number of emails acquired
Reducing Form Field Test
Control Treatment
483 increasein conversion rate
Experiment 4699
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
onlineforlifeorg
Todayrsquos Speakers
Tim KachuriakChief Innovation amp Optimization Officer NextAfter
Jason GSellLaunch Manager Optimizely
Amish S PatelLaunch Manager Optimizely
onlineforlifeorg
6 Steps to Launching Effective Tests on Your Website
Featuring Our Friends From
onlineforlifeorg
Before We Talk About Howto Test Letrsquos Start With
Where to Test What to Test
Three Key Areas to Focus Your Testing
Email Signup Pages Landing Pages Donation Pages
Three Key Areas to Focus Your Testing
Email Signup Pages Landing Pages Donation Pages
What to Test Value PropositionCopyAdding CopyRemoving CopyDifferent Copy
Number of Form FieldsButton Text
Optimizing Email Signup Pages
Email Signup Pages
onlineforlifeorg
How communicating value of an email newsletter impacts email acquisitionExperiment 1621
Communicating Value
Backgroundbull Partner Alliance Defending
Freedombull Tested email acquisition on
the home pagebull Home page receives
~12000 visitors each month with less than 1 giving their email
Control
Communicating ValueTreatment
Treatment Changesbull Added a call to action to the
headlinebull Included value proposition
language identifying the benefits of giving an email
bull Button communicated value
Communicating ValueControl Treatment
441 increase in emails acquired
Communicating Value
Version Conv Rate Relative Diff Stat Confidence
C Simple form 09T1 Value proposition form 13 441 999Key Learning
Every interaction with a visitor is a potential value exchange By communicating the benefits and value to the visitor we increasedconversion by 441
441 increase in emails acquired
Sometimes You Have to Eliminate Copy
Control Treatment
Experiment 2313
262 increase in emails acquired
Itrsquos Not Just Copy But the Right Copy
Control Treatment
295 decrease in emails acquiredExperiment 2606
onlineforlifeorg
How the amount of required information impacts conversionExperiment 289
Impact of Required Form Fields
Backgroundbull Client Hillsdale Collegebull Name acquisition offer for their
Imprimis publicationbull The form historically required both
email and home address so it could be send through mail and online
Control
Impact of Required Form Fields
Tested Elementsbull Reduced the required form
fields by removing address information
bull Removed image of Imprimisissues
Treatment
Treatment
Impact of Required Form Fields
1360 increasein conversion rate
Control
Impact of Required Form Fields
1360 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Full Address 322T1 Email Only 760 1360 1000
Key Learning
By removing the friction associated with the home address fields Hillsdale was able to more than double the number of emails acquired
Reducing Form Field Test
Control Treatment
483 increasein conversion rate
Experiment 4699
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
onlineforlifeorg
6 Steps to Launching Effective Tests on Your Website
Featuring Our Friends From
onlineforlifeorg
Before We Talk About Howto Test Letrsquos Start With
Where to Test What to Test
Three Key Areas to Focus Your Testing
Email Signup Pages Landing Pages Donation Pages
Three Key Areas to Focus Your Testing
Email Signup Pages Landing Pages Donation Pages
What to Test Value PropositionCopyAdding CopyRemoving CopyDifferent Copy
Number of Form FieldsButton Text
Optimizing Email Signup Pages
Email Signup Pages
onlineforlifeorg
How communicating value of an email newsletter impacts email acquisitionExperiment 1621
Communicating Value
Backgroundbull Partner Alliance Defending
Freedombull Tested email acquisition on
the home pagebull Home page receives
~12000 visitors each month with less than 1 giving their email
Control
Communicating ValueTreatment
Treatment Changesbull Added a call to action to the
headlinebull Included value proposition
language identifying the benefits of giving an email
bull Button communicated value
Communicating ValueControl Treatment
441 increase in emails acquired
Communicating Value
Version Conv Rate Relative Diff Stat Confidence
C Simple form 09T1 Value proposition form 13 441 999Key Learning
Every interaction with a visitor is a potential value exchange By communicating the benefits and value to the visitor we increasedconversion by 441
441 increase in emails acquired
Sometimes You Have to Eliminate Copy
Control Treatment
Experiment 2313
262 increase in emails acquired
Itrsquos Not Just Copy But the Right Copy
Control Treatment
295 decrease in emails acquiredExperiment 2606
onlineforlifeorg
How the amount of required information impacts conversionExperiment 289
Impact of Required Form Fields
Backgroundbull Client Hillsdale Collegebull Name acquisition offer for their
Imprimis publicationbull The form historically required both
email and home address so it could be send through mail and online
Control
Impact of Required Form Fields
Tested Elementsbull Reduced the required form
fields by removing address information
bull Removed image of Imprimisissues
Treatment
Treatment
Impact of Required Form Fields
1360 increasein conversion rate
Control
Impact of Required Form Fields
1360 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Full Address 322T1 Email Only 760 1360 1000
Key Learning
By removing the friction associated with the home address fields Hillsdale was able to more than double the number of emails acquired
Reducing Form Field Test
Control Treatment
483 increasein conversion rate
Experiment 4699
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
onlineforlifeorg
Before We Talk About Howto Test Letrsquos Start With
Where to Test What to Test
Three Key Areas to Focus Your Testing
Email Signup Pages Landing Pages Donation Pages
Three Key Areas to Focus Your Testing
Email Signup Pages Landing Pages Donation Pages
What to Test Value PropositionCopyAdding CopyRemoving CopyDifferent Copy
Number of Form FieldsButton Text
Optimizing Email Signup Pages
Email Signup Pages
onlineforlifeorg
How communicating value of an email newsletter impacts email acquisitionExperiment 1621
Communicating Value
Backgroundbull Partner Alliance Defending
Freedombull Tested email acquisition on
the home pagebull Home page receives
~12000 visitors each month with less than 1 giving their email
Control
Communicating ValueTreatment
Treatment Changesbull Added a call to action to the
headlinebull Included value proposition
language identifying the benefits of giving an email
bull Button communicated value
Communicating ValueControl Treatment
441 increase in emails acquired
Communicating Value
Version Conv Rate Relative Diff Stat Confidence
C Simple form 09T1 Value proposition form 13 441 999Key Learning
Every interaction with a visitor is a potential value exchange By communicating the benefits and value to the visitor we increasedconversion by 441
441 increase in emails acquired
Sometimes You Have to Eliminate Copy
Control Treatment
Experiment 2313
262 increase in emails acquired
Itrsquos Not Just Copy But the Right Copy
Control Treatment
295 decrease in emails acquiredExperiment 2606
onlineforlifeorg
How the amount of required information impacts conversionExperiment 289
Impact of Required Form Fields
Backgroundbull Client Hillsdale Collegebull Name acquisition offer for their
Imprimis publicationbull The form historically required both
email and home address so it could be send through mail and online
Control
Impact of Required Form Fields
Tested Elementsbull Reduced the required form
fields by removing address information
bull Removed image of Imprimisissues
Treatment
Treatment
Impact of Required Form Fields
1360 increasein conversion rate
Control
Impact of Required Form Fields
1360 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Full Address 322T1 Email Only 760 1360 1000
Key Learning
By removing the friction associated with the home address fields Hillsdale was able to more than double the number of emails acquired
Reducing Form Field Test
Control Treatment
483 increasein conversion rate
Experiment 4699
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Three Key Areas to Focus Your Testing
Email Signup Pages Landing Pages Donation Pages
Three Key Areas to Focus Your Testing
Email Signup Pages Landing Pages Donation Pages
What to Test Value PropositionCopyAdding CopyRemoving CopyDifferent Copy
Number of Form FieldsButton Text
Optimizing Email Signup Pages
Email Signup Pages
onlineforlifeorg
How communicating value of an email newsletter impacts email acquisitionExperiment 1621
Communicating Value
Backgroundbull Partner Alliance Defending
Freedombull Tested email acquisition on
the home pagebull Home page receives
~12000 visitors each month with less than 1 giving their email
Control
Communicating ValueTreatment
Treatment Changesbull Added a call to action to the
headlinebull Included value proposition
language identifying the benefits of giving an email
bull Button communicated value
Communicating ValueControl Treatment
441 increase in emails acquired
Communicating Value
Version Conv Rate Relative Diff Stat Confidence
C Simple form 09T1 Value proposition form 13 441 999Key Learning
Every interaction with a visitor is a potential value exchange By communicating the benefits and value to the visitor we increasedconversion by 441
441 increase in emails acquired
Sometimes You Have to Eliminate Copy
Control Treatment
Experiment 2313
262 increase in emails acquired
Itrsquos Not Just Copy But the Right Copy
Control Treatment
295 decrease in emails acquiredExperiment 2606
onlineforlifeorg
How the amount of required information impacts conversionExperiment 289
Impact of Required Form Fields
Backgroundbull Client Hillsdale Collegebull Name acquisition offer for their
Imprimis publicationbull The form historically required both
email and home address so it could be send through mail and online
Control
Impact of Required Form Fields
Tested Elementsbull Reduced the required form
fields by removing address information
bull Removed image of Imprimisissues
Treatment
Treatment
Impact of Required Form Fields
1360 increasein conversion rate
Control
Impact of Required Form Fields
1360 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Full Address 322T1 Email Only 760 1360 1000
Key Learning
By removing the friction associated with the home address fields Hillsdale was able to more than double the number of emails acquired
Reducing Form Field Test
Control Treatment
483 increasein conversion rate
Experiment 4699
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Three Key Areas to Focus Your Testing
Email Signup Pages Landing Pages Donation Pages
What to Test Value PropositionCopyAdding CopyRemoving CopyDifferent Copy
Number of Form FieldsButton Text
Optimizing Email Signup Pages
Email Signup Pages
onlineforlifeorg
How communicating value of an email newsletter impacts email acquisitionExperiment 1621
Communicating Value
Backgroundbull Partner Alliance Defending
Freedombull Tested email acquisition on
the home pagebull Home page receives
~12000 visitors each month with less than 1 giving their email
Control
Communicating ValueTreatment
Treatment Changesbull Added a call to action to the
headlinebull Included value proposition
language identifying the benefits of giving an email
bull Button communicated value
Communicating ValueControl Treatment
441 increase in emails acquired
Communicating Value
Version Conv Rate Relative Diff Stat Confidence
C Simple form 09T1 Value proposition form 13 441 999Key Learning
Every interaction with a visitor is a potential value exchange By communicating the benefits and value to the visitor we increasedconversion by 441
441 increase in emails acquired
Sometimes You Have to Eliminate Copy
Control Treatment
Experiment 2313
262 increase in emails acquired
Itrsquos Not Just Copy But the Right Copy
Control Treatment
295 decrease in emails acquiredExperiment 2606
onlineforlifeorg
How the amount of required information impacts conversionExperiment 289
Impact of Required Form Fields
Backgroundbull Client Hillsdale Collegebull Name acquisition offer for their
Imprimis publicationbull The form historically required both
email and home address so it could be send through mail and online
Control
Impact of Required Form Fields
Tested Elementsbull Reduced the required form
fields by removing address information
bull Removed image of Imprimisissues
Treatment
Treatment
Impact of Required Form Fields
1360 increasein conversion rate
Control
Impact of Required Form Fields
1360 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Full Address 322T1 Email Only 760 1360 1000
Key Learning
By removing the friction associated with the home address fields Hillsdale was able to more than double the number of emails acquired
Reducing Form Field Test
Control Treatment
483 increasein conversion rate
Experiment 4699
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
What to Test Value PropositionCopyAdding CopyRemoving CopyDifferent Copy
Number of Form FieldsButton Text
Optimizing Email Signup Pages
Email Signup Pages
onlineforlifeorg
How communicating value of an email newsletter impacts email acquisitionExperiment 1621
Communicating Value
Backgroundbull Partner Alliance Defending
Freedombull Tested email acquisition on
the home pagebull Home page receives
~12000 visitors each month with less than 1 giving their email
Control
Communicating ValueTreatment
Treatment Changesbull Added a call to action to the
headlinebull Included value proposition
language identifying the benefits of giving an email
bull Button communicated value
Communicating ValueControl Treatment
441 increase in emails acquired
Communicating Value
Version Conv Rate Relative Diff Stat Confidence
C Simple form 09T1 Value proposition form 13 441 999Key Learning
Every interaction with a visitor is a potential value exchange By communicating the benefits and value to the visitor we increasedconversion by 441
441 increase in emails acquired
Sometimes You Have to Eliminate Copy
Control Treatment
Experiment 2313
262 increase in emails acquired
Itrsquos Not Just Copy But the Right Copy
Control Treatment
295 decrease in emails acquiredExperiment 2606
onlineforlifeorg
How the amount of required information impacts conversionExperiment 289
Impact of Required Form Fields
Backgroundbull Client Hillsdale Collegebull Name acquisition offer for their
Imprimis publicationbull The form historically required both
email and home address so it could be send through mail and online
Control
Impact of Required Form Fields
Tested Elementsbull Reduced the required form
fields by removing address information
bull Removed image of Imprimisissues
Treatment
Treatment
Impact of Required Form Fields
1360 increasein conversion rate
Control
Impact of Required Form Fields
1360 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Full Address 322T1 Email Only 760 1360 1000
Key Learning
By removing the friction associated with the home address fields Hillsdale was able to more than double the number of emails acquired
Reducing Form Field Test
Control Treatment
483 increasein conversion rate
Experiment 4699
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
onlineforlifeorg
How communicating value of an email newsletter impacts email acquisitionExperiment 1621
Communicating Value
Backgroundbull Partner Alliance Defending
Freedombull Tested email acquisition on
the home pagebull Home page receives
~12000 visitors each month with less than 1 giving their email
Control
Communicating ValueTreatment
Treatment Changesbull Added a call to action to the
headlinebull Included value proposition
language identifying the benefits of giving an email
bull Button communicated value
Communicating ValueControl Treatment
441 increase in emails acquired
Communicating Value
Version Conv Rate Relative Diff Stat Confidence
C Simple form 09T1 Value proposition form 13 441 999Key Learning
Every interaction with a visitor is a potential value exchange By communicating the benefits and value to the visitor we increasedconversion by 441
441 increase in emails acquired
Sometimes You Have to Eliminate Copy
Control Treatment
Experiment 2313
262 increase in emails acquired
Itrsquos Not Just Copy But the Right Copy
Control Treatment
295 decrease in emails acquiredExperiment 2606
onlineforlifeorg
How the amount of required information impacts conversionExperiment 289
Impact of Required Form Fields
Backgroundbull Client Hillsdale Collegebull Name acquisition offer for their
Imprimis publicationbull The form historically required both
email and home address so it could be send through mail and online
Control
Impact of Required Form Fields
Tested Elementsbull Reduced the required form
fields by removing address information
bull Removed image of Imprimisissues
Treatment
Treatment
Impact of Required Form Fields
1360 increasein conversion rate
Control
Impact of Required Form Fields
1360 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Full Address 322T1 Email Only 760 1360 1000
Key Learning
By removing the friction associated with the home address fields Hillsdale was able to more than double the number of emails acquired
Reducing Form Field Test
Control Treatment
483 increasein conversion rate
Experiment 4699
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Communicating Value
Backgroundbull Partner Alliance Defending
Freedombull Tested email acquisition on
the home pagebull Home page receives
~12000 visitors each month with less than 1 giving their email
Control
Communicating ValueTreatment
Treatment Changesbull Added a call to action to the
headlinebull Included value proposition
language identifying the benefits of giving an email
bull Button communicated value
Communicating ValueControl Treatment
441 increase in emails acquired
Communicating Value
Version Conv Rate Relative Diff Stat Confidence
C Simple form 09T1 Value proposition form 13 441 999Key Learning
Every interaction with a visitor is a potential value exchange By communicating the benefits and value to the visitor we increasedconversion by 441
441 increase in emails acquired
Sometimes You Have to Eliminate Copy
Control Treatment
Experiment 2313
262 increase in emails acquired
Itrsquos Not Just Copy But the Right Copy
Control Treatment
295 decrease in emails acquiredExperiment 2606
onlineforlifeorg
How the amount of required information impacts conversionExperiment 289
Impact of Required Form Fields
Backgroundbull Client Hillsdale Collegebull Name acquisition offer for their
Imprimis publicationbull The form historically required both
email and home address so it could be send through mail and online
Control
Impact of Required Form Fields
Tested Elementsbull Reduced the required form
fields by removing address information
bull Removed image of Imprimisissues
Treatment
Treatment
Impact of Required Form Fields
1360 increasein conversion rate
Control
Impact of Required Form Fields
1360 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Full Address 322T1 Email Only 760 1360 1000
Key Learning
By removing the friction associated with the home address fields Hillsdale was able to more than double the number of emails acquired
Reducing Form Field Test
Control Treatment
483 increasein conversion rate
Experiment 4699
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Communicating ValueTreatment
Treatment Changesbull Added a call to action to the
headlinebull Included value proposition
language identifying the benefits of giving an email
bull Button communicated value
Communicating ValueControl Treatment
441 increase in emails acquired
Communicating Value
Version Conv Rate Relative Diff Stat Confidence
C Simple form 09T1 Value proposition form 13 441 999Key Learning
Every interaction with a visitor is a potential value exchange By communicating the benefits and value to the visitor we increasedconversion by 441
441 increase in emails acquired
Sometimes You Have to Eliminate Copy
Control Treatment
Experiment 2313
262 increase in emails acquired
Itrsquos Not Just Copy But the Right Copy
Control Treatment
295 decrease in emails acquiredExperiment 2606
onlineforlifeorg
How the amount of required information impacts conversionExperiment 289
Impact of Required Form Fields
Backgroundbull Client Hillsdale Collegebull Name acquisition offer for their
Imprimis publicationbull The form historically required both
email and home address so it could be send through mail and online
Control
Impact of Required Form Fields
Tested Elementsbull Reduced the required form
fields by removing address information
bull Removed image of Imprimisissues
Treatment
Treatment
Impact of Required Form Fields
1360 increasein conversion rate
Control
Impact of Required Form Fields
1360 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Full Address 322T1 Email Only 760 1360 1000
Key Learning
By removing the friction associated with the home address fields Hillsdale was able to more than double the number of emails acquired
Reducing Form Field Test
Control Treatment
483 increasein conversion rate
Experiment 4699
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Communicating ValueControl Treatment
441 increase in emails acquired
Communicating Value
Version Conv Rate Relative Diff Stat Confidence
C Simple form 09T1 Value proposition form 13 441 999Key Learning
Every interaction with a visitor is a potential value exchange By communicating the benefits and value to the visitor we increasedconversion by 441
441 increase in emails acquired
Sometimes You Have to Eliminate Copy
Control Treatment
Experiment 2313
262 increase in emails acquired
Itrsquos Not Just Copy But the Right Copy
Control Treatment
295 decrease in emails acquiredExperiment 2606
onlineforlifeorg
How the amount of required information impacts conversionExperiment 289
Impact of Required Form Fields
Backgroundbull Client Hillsdale Collegebull Name acquisition offer for their
Imprimis publicationbull The form historically required both
email and home address so it could be send through mail and online
Control
Impact of Required Form Fields
Tested Elementsbull Reduced the required form
fields by removing address information
bull Removed image of Imprimisissues
Treatment
Treatment
Impact of Required Form Fields
1360 increasein conversion rate
Control
Impact of Required Form Fields
1360 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Full Address 322T1 Email Only 760 1360 1000
Key Learning
By removing the friction associated with the home address fields Hillsdale was able to more than double the number of emails acquired
Reducing Form Field Test
Control Treatment
483 increasein conversion rate
Experiment 4699
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Communicating Value
Version Conv Rate Relative Diff Stat Confidence
C Simple form 09T1 Value proposition form 13 441 999Key Learning
Every interaction with a visitor is a potential value exchange By communicating the benefits and value to the visitor we increasedconversion by 441
441 increase in emails acquired
Sometimes You Have to Eliminate Copy
Control Treatment
Experiment 2313
262 increase in emails acquired
Itrsquos Not Just Copy But the Right Copy
Control Treatment
295 decrease in emails acquiredExperiment 2606
onlineforlifeorg
How the amount of required information impacts conversionExperiment 289
Impact of Required Form Fields
Backgroundbull Client Hillsdale Collegebull Name acquisition offer for their
Imprimis publicationbull The form historically required both
email and home address so it could be send through mail and online
Control
Impact of Required Form Fields
Tested Elementsbull Reduced the required form
fields by removing address information
bull Removed image of Imprimisissues
Treatment
Treatment
Impact of Required Form Fields
1360 increasein conversion rate
Control
Impact of Required Form Fields
1360 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Full Address 322T1 Email Only 760 1360 1000
Key Learning
By removing the friction associated with the home address fields Hillsdale was able to more than double the number of emails acquired
Reducing Form Field Test
Control Treatment
483 increasein conversion rate
Experiment 4699
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Sometimes You Have to Eliminate Copy
Control Treatment
Experiment 2313
262 increase in emails acquired
Itrsquos Not Just Copy But the Right Copy
Control Treatment
295 decrease in emails acquiredExperiment 2606
onlineforlifeorg
How the amount of required information impacts conversionExperiment 289
Impact of Required Form Fields
Backgroundbull Client Hillsdale Collegebull Name acquisition offer for their
Imprimis publicationbull The form historically required both
email and home address so it could be send through mail and online
Control
Impact of Required Form Fields
Tested Elementsbull Reduced the required form
fields by removing address information
bull Removed image of Imprimisissues
Treatment
Treatment
Impact of Required Form Fields
1360 increasein conversion rate
Control
Impact of Required Form Fields
1360 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Full Address 322T1 Email Only 760 1360 1000
Key Learning
By removing the friction associated with the home address fields Hillsdale was able to more than double the number of emails acquired
Reducing Form Field Test
Control Treatment
483 increasein conversion rate
Experiment 4699
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Itrsquos Not Just Copy But the Right Copy
Control Treatment
295 decrease in emails acquiredExperiment 2606
onlineforlifeorg
How the amount of required information impacts conversionExperiment 289
Impact of Required Form Fields
Backgroundbull Client Hillsdale Collegebull Name acquisition offer for their
Imprimis publicationbull The form historically required both
email and home address so it could be send through mail and online
Control
Impact of Required Form Fields
Tested Elementsbull Reduced the required form
fields by removing address information
bull Removed image of Imprimisissues
Treatment
Treatment
Impact of Required Form Fields
1360 increasein conversion rate
Control
Impact of Required Form Fields
1360 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Full Address 322T1 Email Only 760 1360 1000
Key Learning
By removing the friction associated with the home address fields Hillsdale was able to more than double the number of emails acquired
Reducing Form Field Test
Control Treatment
483 increasein conversion rate
Experiment 4699
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
onlineforlifeorg
How the amount of required information impacts conversionExperiment 289
Impact of Required Form Fields
Backgroundbull Client Hillsdale Collegebull Name acquisition offer for their
Imprimis publicationbull The form historically required both
email and home address so it could be send through mail and online
Control
Impact of Required Form Fields
Tested Elementsbull Reduced the required form
fields by removing address information
bull Removed image of Imprimisissues
Treatment
Treatment
Impact of Required Form Fields
1360 increasein conversion rate
Control
Impact of Required Form Fields
1360 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Full Address 322T1 Email Only 760 1360 1000
Key Learning
By removing the friction associated with the home address fields Hillsdale was able to more than double the number of emails acquired
Reducing Form Field Test
Control Treatment
483 increasein conversion rate
Experiment 4699
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Impact of Required Form Fields
Backgroundbull Client Hillsdale Collegebull Name acquisition offer for their
Imprimis publicationbull The form historically required both
email and home address so it could be send through mail and online
Control
Impact of Required Form Fields
Tested Elementsbull Reduced the required form
fields by removing address information
bull Removed image of Imprimisissues
Treatment
Treatment
Impact of Required Form Fields
1360 increasein conversion rate
Control
Impact of Required Form Fields
1360 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Full Address 322T1 Email Only 760 1360 1000
Key Learning
By removing the friction associated with the home address fields Hillsdale was able to more than double the number of emails acquired
Reducing Form Field Test
Control Treatment
483 increasein conversion rate
Experiment 4699
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Impact of Required Form Fields
Tested Elementsbull Reduced the required form
fields by removing address information
bull Removed image of Imprimisissues
Treatment
Treatment
Impact of Required Form Fields
1360 increasein conversion rate
Control
Impact of Required Form Fields
1360 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Full Address 322T1 Email Only 760 1360 1000
Key Learning
By removing the friction associated with the home address fields Hillsdale was able to more than double the number of emails acquired
Reducing Form Field Test
Control Treatment
483 increasein conversion rate
Experiment 4699
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Treatment
Impact of Required Form Fields
1360 increasein conversion rate
Control
Impact of Required Form Fields
1360 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Full Address 322T1 Email Only 760 1360 1000
Key Learning
By removing the friction associated with the home address fields Hillsdale was able to more than double the number of emails acquired
Reducing Form Field Test
Control Treatment
483 increasein conversion rate
Experiment 4699
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Impact of Required Form Fields
1360 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Full Address 322T1 Email Only 760 1360 1000
Key Learning
By removing the friction associated with the home address fields Hillsdale was able to more than double the number of emails acquired
Reducing Form Field Test
Control Treatment
483 increasein conversion rate
Experiment 4699
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Reducing Form Field Test
Control Treatment
483 increasein conversion rate
Experiment 4699
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
What to Test Design Elements
HeaderLayoutSequence
CopyCalls-to-Action
Optimizing Landing Pages
Landing Pages
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
onlineforlifeorg
How clarifying the design of the page header affects email acquisition
Experiment 2684
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Headline Placement Test
Backgroundbull Client Hillsdale Collegebull Email acquisition campaign
with traffic being driven from Facebook
bull Control uses a traditional page header that contains the headline for the landing page
Control
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull New Headline
Treatment 1
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Headline Placement Test
Tested Elementsbull Removed the graphical
Headerbull Identical Headline as control
Treatment
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Headline Placement Test
Control Treatment 1
66 increasein conversion rate
Treatment 2
92 increasein conversion rate
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Headline Placement Test
92 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 375Treatment 1 400 66 963Treatment 2 410 92 994
Key Learning
By removing the header graphic and moving the headline into the the content area we created a tighter connection the value proposition
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
onlineforlifeorg
How Increasing Landing Page Congruence Reduced Cost Per SubscriberExperiment 314
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Increasing Landing Page CongruenceBackgroundbull Client Hoover Institutionbull Name acquisition for
Strategika newsletterbull Two-column layoutbull Branded headerbull Clear calls to action
Control
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
The problemControlThe ad
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Increasing Landing Page CongruenceTreatmentTreatment Changes
bull Matched background image to the ad to increase congruence
bull Took the brand (which was unfamiliar to the visitor) out of the headline
bull Highlighted credibility factors instead of showing unknown contributors
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Increasing Landing Page CongruenceControl Treatment
392 increasein conversion rate
86 decreasein costsubscriber
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Increasing Landing Page Congruence
Key Learning
The site visitor is more likely to respond when the ad and the landingpage maintain a consistent visual experience Additionally when selling anunfamiliar brand leading with the passion point can increase conversion
392 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Branded Page 250T1 Congruent Page 348 392 1000
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
onlineforlifeorg
How reordering the elements of a landing page increased name acquisition rate by 108Experiment ID 1692
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Reordering elements increased conversion
Backgroundbull Client Stanford Graduate
School of Businessbull Email acquisition page for
eBook offerbull Three-column layout
mirrored branding throughout the site
Control
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Reordering elements increased conversion
Treatment
Treatment Changesbull Changed headline to convey
valuebull Gave copy contextual
placement near the formbull Moved email acquisition form
into eyepathbull Moved image and credibility
indicators to the right column as supporting content
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Reordering elements increased conversion
Control Treatment
108 increasein conversion rate
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Reordering elements increased conversion
Version Conv Rate Relative Diff Stat ConfidenceC Three-column page 387
T1 Optimized page 429 108 989
Key Learning
Simply matching branding throughout the site is not enough Each landing page must be optimized maximize perceived value and minimizeperceived cost
108 increase in conversion rate
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Re-Ordering Thought Sequences
Control Treatment
77 increasein conversion rateExperiment ID 2472
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
onlineforlifeorg
How clarifying the process-level value proposition affects email acquisition
Experiment 833
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Process-Level Value Proposition Test
Backgroundbull Client Good of Allbull Email acquisition campaign
with traffic being driven from Facebook
bull Call-to-action focuses on what I need to do instead of what I can get
Control
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Process-Level Value Proposition Test
Tested Elementsbull Copy addresses reader as a
ldquofellow world-changerrdquobull Call-to-action focuses on
what I can get and when I can get it
Treatment
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
onlineforlifeorg
Process-Level Value Proposition TestControl Treatment
1337 increasein conversion rate
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Process-level Value Proposition Test
1337 increase in names acquired
Version Conv Rate Relative Diff Stat Confidence
Control 13Treatment 31 1337 980Key Learning
By addressing the reader as a fellow ldquoworld-changerrdquo andcommunicating the value of the offer rather than the actionrequired to receive the offer the conversion rate improved by 1337
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
onlineforlifeorg
How visitor-focused language impacts conversionExperiment 986
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Backgroundbull Client Hillsdalebull The sign up page for the new
course offering The Federalist
bull Internal logic was to use ldquoEnroll forhelliprdquo since it was Hillsdalersquos course
Control
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentTreatment
Treatment Changesbull Changed the call-to-action
above the signup formbull Remaining copy all stayed
the same
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
ldquoEnrollrdquo vs ldquoActivaterdquo ExperimentControl Treatment
315 increasein conversion rate
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
ldquoEnrollrdquo vs ldquoActivaterdquo Experiment
Version Conv Rate Relative Diff Stat Confidence
C ldquoEnrollrdquo 265T1 ldquoActivaterdquo 349 315 999Key Learning
Making the visitors ldquoEnrollrdquo created more mental friction given the perceived work that would be required ldquoActivaterdquo has the perceptionof a one-step process which created a 315 increase in conversion
315 increase in conversion rate
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
What to Test Radical RedesignsContent Elements
HeadlinesCopyVideo vs Text
Form Design
Optimizing Landing Pages
Donation Pages
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
onlineforlifeorg
How the force of the value proposition can be altered with a radical redesignExperiment 3793
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Impact of Radical Redesign
Backgroundbull Client Texas State Historical
Associationbull Primary donation page for
TSHAOnlineorgbull Four different membership
options presentedbull Copy is not specific and
makes unclear and unsubstantiated claims
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Impact of Radical Redesign Treatment
Treatment Changesbull Radical redesign with long
form copy to appropriately convey the value proposition
bull Added copy outlining reasons why visitor should join the TSHA as a member
bull Included evidentials to back up each claim
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
bull Headline arrests attention by conveying an exclusive offer
bull First paragraph focuses on primary value proposition
Impact of Radical Redesign
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
The value proposition is reinforced throughout the copy
Impact of Radical Redesign
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
bull ldquoBenefitsrdquo are used as incentives to move members to higher levels
bull Lowest level named ldquoBasicrdquo to de-incentivize signups at this level
Impact of Radical Redesign
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Impact of Radical Redesign Control Treatment
1465 increasein conversion rate
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Impact of Radical Redesign
Version Conv Rate Relative Diff Stat ConfidenceC Short Form 11
T1 Radical Redesign LF 26 1465 100
Key Learning
A radically different treatment is sometimes needed to get big increasesin conversion By altering the page to more forcefully communicate acompelling argument that highlighted the exclusivity of membership wewere able to increase donations by 1465
1465 increase in conversion rate
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
onlineforlifeorg
How donor-centric headline affects donation conversion rateExperiment 1039
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Clarity Trumps PersuasionControl Treatment
367 increasein Donations
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
onlineforlifeorg
How copy on a donation page affects the force of the value propositionExperiment ID 111
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Power of the Value Proposition
Backgroundbull Client Senator John Cornynbull Primary donation page for
JohnCornyncombull Minimal copy on the page
Control
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Value Proposition Question
ldquoIf I am your ideal donor why should I give to you rather than some other organization or
no one at allrdquo
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Power of the Value PropositionTreatment
Treatment Changesbull Changed headline to convey
valuebull Added copy outlining reasons
why donors support the Senator
bull Included evidentials to back up each claim
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Power of the Value PropositionControl Treatment
2581 increasein Donations
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Power of the Value Proposition
Version Conv Rate Relative Diff Stat ConfidenceC No Value Prop 011
T1 Value Prop on Page 038 2581 963
Key Learning
There is a danger in assuming that your potential donor firmly graspsyour organizationrsquos unique value proposition By simply adding copyto the donation page to communicate value conversion increased byby 258
2581 increase in conversion rate
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Impact of VideoControl Treatment
203 increasein conversion rate
Experiment ID 3970
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
onlineforlifeorg
How does form layout affect friction on a donation formExperiment 1007
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
onlineforlifeorg
Overview CaringBridgeorg CaringBridge is a website where people in a time of need can securely share updates photos and videos connecting with friends and family who care and want to help
Over 500000 people per day access this web site
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Donation Form Friction
Backgroundbull Client Caringbridgebull Primary donation form on
the websitebull Default settings of the
payment form had vertically aligned fields
Control
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Donation Form FrictionTreatment
Treatment Changesbull Grouped similar fields
together to shorten donation form
bull As a result of the grouping reduced the width of several fields
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Donation Form FrictionControl Treatment
394 increasein conversion rate
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Donation Form Friction
Key Learning
Even though the horizontal field layout required the same amount of data it created the perception that less was required thereby reducing friction and creating an increase of 39 This is a great reminder that friction doesnrsquot exist on the page it lives in the mind
394 increase in conversion rate
Version Conv Rate Relative Diff Stat Confidence
C Vertically Aligned 52T1 Grouped Fields 72 394 1000
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
onlineforlifeorg
How a modern design trend impacts conversion on a donation pageExperiment 3552
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Modern Design Trend Experiment
Backgroundbull Partner Hillsdale Collegebull Main Donation Page formbull Used traditional design
styling and static field labels
Control
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Modern Design Trend ExperimentTreatment
Treatment Changesbull Modernized the form by
implementing current design trends
bull Minimized the visual size of the form by removing field labels
bull Implemented new technique ldquoadaptive placeholdersrdquo
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Adaptive Placeholders
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Modern Design Trend ExperimentControl Treatment
697 decreasein donations
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Modern Design Trend Experiment
Version Conv Rate Relative Diff Stat Confidence
C Traditional form 30
T1 Modern Form 091 -697 100
Key Learning
Beware of modern design fads Test everything and donrsquot makedecisions based on whatrsquos popular make decisions on what works
697 decrease in donations
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Optimization Is A Journey
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Optimization Is A Journey
Where
What
How
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
1 How Optimizely works
2 How to create an experiment in 6 steps
3 What to do next
In todayrsquos training yoursquoll learnhellip
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
charity water
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
charity water
Hypothesis
Making the four CTArsquos stand
out will lead to more clicks on the donation buttons
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
charity water
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Experiment Ideas
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Experiment IdeasExperiment 1 Remove content from the headline to eliminate distractions and clutter
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Experiment Ideas
Experiment 2 Change the CTA text to create a clearer message
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Experiment Ideas
Experiment 3 Swap the ldquoGIVE MONTHLYrdquo and ldquoGIVE ONCErdquo sections to improve monthly donors
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
to a Successful Experiment
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Make the Variations What are you testing
1
6 Key Steps
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Make the Variations What are you testing
1
6 Key Steps
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Step 1 What are you testing
Visual Editor
bull Non-technical users
bull Quick updates wo deploying code
bull Copytext changes
bull Image swap
bull Hideremove content
bull Rearrange assets
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Code Editor
bull Front-end developers
bull Add custom jQuery and JavaScript
bull See and edit all your variation code
Step 1 What are you testing
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Make the Variations
Set Goals Why are you testing How will you measure success
1
2
6 Key Steps
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Types of Goals
Click Pageview Custom Event
Step 2 What to Track
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
How should charity water measure success
Experiment Changing the CTA text to create a clearer message which will lead to more ldquodonationrdquo clicks
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Step 2 What to Track
Types of Goals
Click
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Pageview
Step 2 What to Track
Types of Goals
HOMEPAGE
VIEW PROJECTS
VIEW DONATIONS
ADD TO CART
CHECKOUT
$$$
browse abandonment
cart abandonment
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Custom Event
Step 2 What to Track
Types of Goals
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Make the Variations
Set Goals
Set URL Targeting Where (which pages) are you testing
1
2
3
6 Key Steps
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Step 3 Where to Test
How do I run an experiment on a single page like a home
page or a landing page
Simple Match
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Step 3 Where to Test
How do I run an experiment on a page where a queryhash parameter changes the user experience such
as a single-page application
Exact Match
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Step 3 Where to Test
How do I run an experiment on multiple pages such as my entire
sitersquos global navigation all product pages or a check-out funnel
Substring Match
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Make the Variations
Set Goals
Set URL Targeting
Set Audiences Who sees the experiment
1
2
3
4
6 Key Steps
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Step 4 Who Sees the Experiment
Audiences - Target based on
bullAd Campaigns
bullTraffic Source
bullQuery Parameters
bullLocation
bullCookies
bullDevice Type
bullNewReturning Session
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation How many visitors are included How are they split
1
2
3
4
5
6 Key Steps
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Step 5 How Many Visitors
Inout of the experiment
Among variations within the experiment
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test Does it work the way you want it to
1
2
3
4
5
6
6 Key Steps
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Var Code
Step 6 Quality Assurance
Issues to check for
Does your variation look the way it should
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Var Code
Step 6 Quality Assurance
Timing
Issues to check for
Does your variation show up when it should without page flashing
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Var Code
Step 6 Quality Assurance
Timing Targeting
Issues to check for
Does your experiment appear on the right pages for the right audiences
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Var Code
Step 6 Quality Assurance
Timing Targeting Goals
Issues to check for
Do your goals fire when they should and capture accurate data
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Make the Variations
Set Goals
Set URL Targeting
Set Audiences
Set Traffic Allocation
Preview and Test
1
2
3
4
5
6
6 Key Steps
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
Whatrsquos Nextbull Next Steps
bull Build your first experiment
bull How to review todayrsquos discussion bull 6 Key Steps to Build an Experiment Knowledge Base article
bull Optimizely Academy - Foundations of Web AB Testing
bull Do you have questions bull Share in the Community
bull Support - optimizelycomsupport
QampA
QampA