New Product DevelopmentNew Product DevelopmentMKTG 4320.002MKTG 4320.002

Dr. Audhesh PaswanDr. Audhesh Paswan

Spring 2006Spring 2006

M 2:00-4:50 PM (BUSI 330)M 2:00-4:50 PM (BUSI 330)

DML-COBA

Analytical Approaches: Analytical Approaches: Introduction And Perceptual MappingIntroduction And Perceptual Mapping

Ch - 6Ch - 6

What are Analytical Attribute What are Analytical Attribute Techniques?Techniques?

Basic idea: products are made up of attributes -- a future product change must involve one or more of these attributes.

Three types of attributes: features, functions, benefits.

Theoretical sequence: feature permits a function which provides a benefit.

Gap AnalysisGap Analysis Determinant gap map (produced

from managerial input/judgment on products)

Attribute Ratings perceptual gap map (based on attribute ratings by customers)

Overall Smilarity perceptual map (based on overall similarities ratings by customers)

A Determinant Gap MapA Determinant Gap Map

1 2 3 .... Options .... X Ideal1 2 3 .... Options .... X Ideal1122 .. .. .. .. . . .. ..1515

Attr

ibut

esA

ttrib

utes

Respo

nden

ts

Respo

nden

ts

1122 .. ..

700700..

A Data CubeA Data Cube

Rate each brand you are familiar with on each of the following:

Disagree Agree1. Attractive design 1..2..3..4..5 2. Stylish 1..2..3..4..5 3. Comfortable to wear 1..2..3..4..5 4. Fashionable 1..2..3..4..5 5. I feel good when I wear it 1..2..3..4..5 6. Is ideal for swimming 1..2..3..4..57. Looks like a designer label 1..2..3..4..58. Easy to swim in 1..2..3..4..59. In style 1..2..3..4..5 10. Great appearance 1..2..3..4..5 11. Comfortable to swim in 1..2..3..4..5 12. This is a desirable label 1..2..3..4..5 13. Gives me the look I like 1..2..3..4..5 14. I like the colors it comes in 1..2..3..4..5 15. Is functional for swimming 1..2..3..4..5

Obtaining Customer PerceptionsObtaining Customer Perceptions

1

1.5

2

2.5

3

3.5

4

4.5

5

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Snake Plot of PerceptionsSnake Plot of Perceptions ( (Three BrandsThree Brands))

Aqualine

Islands

Sunflare

Attributes

Ratings

Data Reduction Using Multivariate Data Reduction Using Multivariate AnalysisAnalysis

Factor Analysis Reduces the original number of attributes to

a smaller number of factors, each containing a set of attributes that “hang together”

Cluster Analysis Reduces the original number of respondents

to a smaller number of clusters based on their benefits sought, as revealed by their “ideal brand”

Factor Eigenvalue Percent VarianceExplained

1 6.04 40.32 3.34 22.33 0.88 5.94 0.74 4.95 0.62 4.26 0.54 3.67 0.52 3.58 0.44 3.09 0.40 2.7

05

1015202530354045

1 2 3 4 5 6 7 8 9 No. of Factors

Perc

ent V

aria

nce

Expl

aine

d

The Scree

Selecting the Number of FactorsSelecting the Number of Factors

Attribute Factor 1 --“Fashion”

Factor 2 --“Comfort”

1. Attractive design .796 .0612. Stylish .791 .0293. Comfortable to wear .108 .7824. Fashionable .803 .0775. I feel good when I wear it .039 .7296. Is ideal for swimming .102 .8337. Looks like a designer label .754 .0598. Easy to swim in .093 .7939. In style .762 .12310. Great appearance .758 .20811. Comfortable to swim in .043 .75612. This is a desirable label .807 .08213. Gives me the look I like .810 .05514. I like the colors it comes in .800 .06115. Is functional for swimming .106 .798

Factor Loading MatrixFactor Loading Matrix

Attribute Factor 1 --“Fashion”

Factor 2 --“Comfort”

1. Attractive design 0.145 -0.0222. Stylish 0.146 -0.0303. Comfortable to wear -0.018 0.2134. Fashionable 0.146 -0.0175. I feel good when I wear it -0.028 0.2016. Is ideal for swimming -0.021 0.2277. Looks like a designer label 0.138 -0.0208. Easy to swim in 0.131 0.2169. In style -0.021 -0.00310. Great appearance 0.146 0.02111. Comfortable to swim in -0.029 0.20812. This is a desirable label 0.146 -0.01613. Gives me the look I like 0.148 -0.02414. I like the colors it comes in 0.146 -0.02215. Is functional for swimming -0.019 0.217

Sample calculation of factor scores: From the snake plot, the mean ratings of Aqualine on Attributes1 through 15 are 2.15, 2.40, 3.48, …, 3.77. Multiply each of these mean ratings by the correspondingcoefficient in the factor score coefficient matrix to get Aqualine’s factor scores. For example, on Factor 1, Aqualine’s score is (2.15 x 0.145) + (2.40 x 0.146) + (3.48 x -0.018) + … + (3.77 x -0.019)= 2.48. Similarly, its score on Factor 2 can be calculated as 4.36. All other brands’ factor scores are calculated the same way.

Factor Scores MatrixFactor Scores Matrix

Aqualine

Islands

Splash

Molokai

Sunflare

Gap 1

Gap 2

Fashion

Com

fort

The AR Perceptual MapThe AR Perceptual Map

Aqualine Islands Sunflare Molokai SplashAqualine X 3 9 5 7Islands X 8 3 4Sunflare X 5 7Molokai X 6Splash X

Dissimilarity MatrixDissimilarity Matrix

Aqualine

Islands

Splash

Molokai

SunflareCom

fort

Fashion

The OS Perceptual MapThe OS Perceptual Map

AR Methods OS MethodsInput Required

Brand ratings on specific attributes Overall similarity ratingsAttributes must be pre-specified Respondent uses own judgment of similarity

Analytic Procedures Commonly UsedFactor analysis; multiple discriminant analysis Multidimensional scaling (MDS)

Graphical OutputShows product positions on axesAxes interpretable as underlying dimensions(factors)

Shows product positions relative to each otherAxes obtained through follow-up analysis or mustbe interpreted by the researcher

Where UsedSituations where attributes are easily articulated orvisualized

Situations where it may be difficult for therespondent to articulate or visualize attributes

Source: Adapted from Robert J. Dolan, Managing the New Product Development Process: Cases and Notes(Reading, MA: Addison-Wesley, 1993), p. 102.

Comparing AR and OS MethodsComparing AR and OS Methods

Failures of Gap AnalysisFailures of Gap Analysis Input comes from questions on how brands

differ (nuances ignored) Brands considered as sets of attributes;

totalities, interrelationships overlooked; also creations requiring a conceptual leap

Analysis and mapping may be history by the time data are gathered and analyzed

Acceptance of findings by persons turned off by mathematical calculations?