o

I

U)

IUI

UwU,

Uc’JUI

0I.U0

I ONLINE ADVERTISING

Data from several UMSL campaigns that usedbranded search advertising (i.e., the purchase of searchkeywords preceded by an organization’s brand name,UMSL summer school, for example) convinced us tostop investing in this segment of search engine marketing. Still, the decision made us nervous. Many recognizable brands, colleges and universities among them,actively engage in the practice. Who were we to contradict the conventional wisdom of a multibilliondollar industry?

ARE CLICKS WORTH THE COST?We market many things at UMSL, but two of ourmost important efforts are the enrollment marketingcampaigns for our summer sessions and UMSL Dayopen houses, which occur in the spring and fall. Inboth cases, attendance is our goal, but registrations areour deliverables. We use Google Analytics to track webtraffic for the open houses at our UMSL Day microsite(www.umslday.com) through to conversion, which inthis case means registering to attend. The same goes

Web Images Maps Shopping News More Search toots

About 1510000,000 results (0.18 seconds) Branded searchAd related to amazon® resultAmazon,comlb) Official Sitesw.amazon.com/Huge Selection and Amazing Prices. Free Shipping on Orders Over 525

Amazon.com: Online Shopping for Electronics, Apparel, Computerswww.amazon.com/Online retailer of books movies, music and games along with electronics, toys, apparel,sports, tools, groceries and general home and garden items. Region 1

Books Shop by DepartmentHolly Slack talks to Amazon’s Seira Shop by Department Search. All. AllWilson about her new book, The ... Departments, Amazon

Kindle eBooksLooking for your neat great ebook?Over one million ebooks, free

MusicBrowse best sellers, new releases,deals, vinyl records, MP3s, and

Today’s DealsDaily Deals - Deals - Deals &Bargains - Digital Deals . Outlet

for summer school (www.umsl.edu/summer.), exceptapplications are the conversion we measure here.In late 2010 we viewed branded search ads as just

another marketing channel and considered paidsearch, also known as search engine marketing, a strategy that complements traditional media advertising.Accordingly, we made room for paid search in ourmarketing budget by reducing the amount we spenton traditional media.Ahead of the spring UMSL Day in March 2011,

we designed a segmented and targeted multimediacampaign using outdoor, radio, print, and onlinevideo advertising. When we began implementing it inlate January 2011, it was the first time we added paidsearch to our campaign marketing mix. All thesechannels pointed people to the UMSL Day microsite,where our initial results seemed amazing. Hits fromour paid search ads, which used the branded keywordsUMSL Day, accounted for 75 percent of registrations.We started to think it was time to call the caterer andorder more food; we were going to have a crowd.Fortunately, for credibility’s sake, we believe in

restrained optimism and took a wait-and-see approach.The spring 2011 UMSL Day reached its attendancecapacity several days before the event, but the percentage of attendees rose only 2 percent from theprevious fall.We kept these results in mind as we began a similar

multichannel marketing campaign for the 2011 summer session, which had a comparable outcome. Paidsearch hits for the branded keywords UMSL summerschool indicated enrollment could increase 10 percent.We thought it might be time to call the deans andadd class sections, but thankfully, we waited. Summersession enrollment did rise, but not by the numbersour paid search hits had projected.The campaigns, as measured by registrations and

enrollment, were successful, but the analytics wereconfusing. In both cases, we had a high conversionpercentage coming directly from paid search. Weknew the reach, frequency, and circulation numbersfor our traditional media advertising were strongand that the campaign’s creative component wasexcellent. But as we increased our advertising ontraditional media channels, traffic going directly toour microsites increased only slightly. At the sametime, however, microsite hits coming from our paidsearch ads skyrocketed—and so did our costs.

Each time someone clicked on our ads, we had topay a certain amount. Paid search success can be adouble-edged sword. More clicks can lead to moreconversions but also produce a budget-busting bill.(We’ve exceeded our UMSL Day paid search budgetby as much as $2,000 in the closing days of a campaign.) Keyword prices can range anywhere from afew cents to $50 or more, depending on the termsselected and the placement of your ad relative to anykeyword competitors. For example, the terms UMSL

Day or UMSL summer school cost us less than $5 perclick and around $20 per conversion.We could see a pattern, but we couldn’t explain it.

We experimented with paid search ad frequency, campaign duration, and daily budget for the fall UMSLDay in November 2011 and the spring UMSL Dayin March 2012. These campaigns clarified the reasons for the increase in hits from paid search. Whenwe added to our traditional media advertising, suchas increasing a radio ad buy, our paid search hitsspiked—and so did our pay-per-click bill. When wereduced our ad frequency on traditional media, wereceived fewer hits from paid search and organicsearch. When we cut back on paid search, we earnedfewer paid search hits but gained more organic hits.For the same budget, we saw a small drop in the number of registrations, but the percentage of attendeesincreased 10 percent. We had discovered a correlationand uncovered an incredible flaw in branded search.But what was the root cause?

WHAT BRANDS CAN LEARN FROM EBAYUltimately, the answer came to us when Ron Gossenpresented the campaign outcomes as a case study tothe graduate-level marketing class he taught in spring2012. We realized we weren’t finding new prospects;we were simply providing an easy—and expensive-pathway to conversion for people who were alreadyinterested. There’s no need to remember a websiteaddress from a radio ad or billboard when you can

Google UMSL Day or UMSL summer school. And ifthe top result is a paid ad, why not click on it? Peoplechose the advertisement’s link because it was there.We had created a registration utility, not a new marketing channel. What was efficient for prospects wasa costly practice for us.We tested these results in successive UMSL Day

and summer session campaigns and achieved thesame outcome. For UMSL, a brand with near universal name recognition in our market, branded search

was serving an informational purpose rather than amarketing function. Buying branded keywords was awaste of our marketing budget because we capturedprospects already exposed to our advertising-peoplewho would have found us through organic search.Our data told us that branded search wasn’t effec

tive, but we couldn’t find any research to supportour conclusion. Then in March 2013, our empiricalfindings were supported by an eBay study, ConsumerHeterogeneity and Paid Search Effectiveness: A LargeScale Field Experiment. EBay has more than 170 million keywords, so investing in a large research project(bit. ly/ebay-study) to test the efficacy of paid searchadvertising and the performance of branded versusnon-branded keywords was in the online auction company’s interest. After suspending its branded keywordadvertising on Google in 65 large U.S. metropolitanmarket areas, eBay determined that branded keywordsearches are of little use for well-known brands.When eBay stopped running branded search ads,

it didn’t lose website hits, just hits from paid search.Consumers found the same link through organicsearch results. These findings told us that if we hadeliminated branded search from our earlier campaigns,our microsites likely would have received roughly thesame amount of traffic from organic search resultssince we hold the top ranking for the terms UMSLDay and UMSL summer school. People will find us—and it won’t cost us any money. (This is why it paysto invest in search engine optimization.)

DOUBLE TROUBLE: A March 2013study by eBay found that brandedkeyword advertising for recognizable brands such as Amazon.comis ineffective land potentially veryexpensive) because Amazon.com’swebsite will generally be the toporganic, or free, search engineresult.

Go. gIe amazon

People chose the advertIsement’s link because it was there. wWhat was efficient for prospects was a costly practice for us.

,/‘ganic (tree)search engine

result

Your AccountView Gift Card Balance Apply a GiftCard to Your Account

More results from amazon.com a

News for amazonAmazon says It won’t launch a free phone

z USA TODAY -3 hours ago- Amazon.corn, the world’s largest retailer has been rumored to be working

on a smartphone of its own for yearn. But the company says there’s noThnes of Ind,aRemain Calm: The Free Amazon Smartohone Isn’t HappeningTIME - by Jared Newman - 23 hours agoAma,nn will rennrtaHlv note awe,, its cmertnhnnn One free

50 0 C T 0 B E R 2 0 1 3 L CASECURRENTS 51

ONLINE ADVERTISING

DON’T WRITE OFF PAID SEARCH ENTIRELYWhile we agree with eBay’s branded search findings,we have a different perspective on non-branded keywords. EBay’s results in this area may have been statistically insignificant, but the non-branded segmentof paid search advertising plays an important role inmarketing UMSL’s degree programs and campustours. It’s delivered a positive return on investment.We use non-branded keywords, such as nursing

school and criminaljustice degree, to promote some ofour highly ranked academic offerings that align withpopular search engine categories. But we also usenon-branded search to market highly competitiveofferings like our MBA program. Expenses add up onthis side too: Using MBA as a keyword can cost us asmuch as $15 per click and $300 for a conversion.But the marketplace for graduate business programsis particularly crowded, so we need to be where prospective students are looking.

The eBav study showed that consumers responddifferently to branded and non-branded search advertising. These findings confirmed our experience aswell. When people see a branded search ad, they generally choose the top result. But people tend to take acloser look at the content of non-branded search adsto find the right result. Therefore, describing yourproduct or program in a way that distinguishes itfrom your competitors—known as a unique sellingproposition—is essential.On average, having UMSL’s non-branded ads

appear in the second ad position on the top of thesearch engine results page is an ideal placement. Ourad displays prominently enough to be easily noticedbut we pay less per click than the competitor whosead ranks above ours. We ensure we’re reaching likelyprospects by using geographic targeting to limit theappearance of our ads to Missouri and Illinois.Our non-branded search ads have been successful

and cost-effective because we invest in creative andcompelling advertising messages that differentiate ourprograms from the competition in ways that leadpeople to click on them. Ideally, this click will leadto a conversion.Although we have seen positive results from our

use of non-branded search, the eBay study remindedus to proceed with caution. Not every degree program has an effective unique selling proposition. Ourchallenge is to improve upon our conversion resultsby investing in strong advertising messages for thosethat do (and to ensure our search engine optimizationis top-notch). Unlike eBay, UMSL doesn’t managemillions of non-branded keywords, so we can affordto put in the time and effort to creatively differentiateand promote our programs through paid search advertising (especially now that we’ve eliminated brandedsearch and shifted those resources to a more efficientand effective use of our marketing dollars).

As higher education marketing rapidly evolves, wewill need to continually reassess our marketing mix.Some people may compare this to a never-endinggame ofWhac-A-Mole, but we prefer the example ofSisyphus and his rock. It’s a bit more dignified. U

Ron Gossen is the senior associate vice chancellor for university mar

keting and communications at the University of Missouri—St. Louis,

where he also teaches in the College of Business Administration, ion

Hinderliter is UMSL’s assistant director of communications—web and

digital marketing manager.

CASE is grateful to thesesupporters for helping tomake the 2013 Conferencefor Community CollegeAdvancement a success.

Save the date for next year’sconference in Sacramento, Calif.:Sept. 17—19, 2014

The Conferencfor ‘ ‘ -

_

TbankYou!GOLD pr

BEACON POINTE

ADVISOR S

Conmoi

FOUNDATIONfor CAIJFORNIAV COMMUNITY COIIEGES

L1I

SILVER SPONSORS

Fund Evaluation Group

Payden&RygelInvestment Management

BRONZE SPONSOR

HAR RI S%CDflflECT

HOSTNetwork of CaliforniaCommunity College

(‘ Foundations

You know who you are. You’re resilient, passionate,I—I re S to curious, smart. You see the bigger picture and standup for change.

From Creosote’s beginning, we’ve rebelled againstt h e re b e I . the business-as-usual approach to marketing andbranding in education. We stand firmly at thecrossroads of creative storytelling and technology—Those who challenge the developing more effective branding strategies at a

status quo. Who never follow better price.a crowd. Who aren’t afraid of

So come on. Rebel. Innovate at a time whenthe unknown. Who think new. demographic data and economic reality demandit. Change. Create the greatest opportunities for asuccessful and sustainable future.

Defy the conventions that no longer have relevancein our market.

re ‘n ir admissions advancement branding researchL cc$’u’raf ects CreosoteAffects.com • 301-447-2EDU [2338]

52 0 C T 0 B E R 2 0 1 3 CASECURRENTS 53