16-1 McGraw-Hill/Irwin © 2003 The McGraw-Hill Companies, Inc., All Rights Reserved. 26-2 CHAPTER SIX ANALYTICAL ATTRIBUTE APPROACHES: INTRODUCTION AND PERCEPTUAL MAPPING 36-3 What are Analytical Attribute 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. 46-4 Gap Analysis • Determinant gap map (produced from managerial input/judgment on products) • AR perceptual gap map (based on attribute ratings by customers) • OS perceptual map (based on overall similarities ratings by customers) 56-5 A Determinant Gap Map Figure 6.2 66-6 A Data Cube 2. 1 Attributes 1 2 . . . . . . . 15 Figure 6.3 700 . . 1 2 3 .... Options .... X Ideal 76-7 Obtaining Customer Perceptions Figure 6.4 Rate each brand you are familiar with on each of the following: Disagree 1. Attractive design 2. Stylish 3. Comfortable to wear 4. Fashionable 5. I feel good when I wear it 6. Is ideal for swimming 7. Looks like a designer label 8. Easy to swim in 9. In style 10. Great appearance 11. Comfortable to swim in 12. This is a desirable label 13. Gives me the look I like 14. I like the colors it comes in 15. Is functional for swimming Agree 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 86-8 Snake Plot of Perceptions (Three Brands) Figure 6.5 Ratings 5 4.5 4 Aqualine 3.5 Islands 3 2.5 Sunflare 2 1.5 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Attributes 96-9 Data Reduction Using Multivariate Analysis • 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” Selecting the Appropriate Number of Factors Percent Variance Explained 6-10 10 45 40 35 30 25 20 15 10 5 0 The Scree 1 2 3 4 5 Factor Eigenvalue 1 2 3 4 5 6 7 8 9 6.04 3.34 0.88 0.74 0.62 0.54 0.52 0.44 0.40 6 7 8 9 Percent Variance Explained 40.3 22.3 5.9 4.9 4.2 3.6 3.5 3.0 2.7 No. of Factors Figure 6.6 6-11 11 Factor Loading Matrix Attribute 1. Attractive design 2. Stylish 3. Comfortable to wear 4. Fashionable 5. I feel good when I wear it 6. Is ideal for swimming 7. Looks like a designer label 8. Easy to swim in 9. In style 10. Great appearance 11. Comfortable to swim in 12. This is a desirable label 13. Gives me the look I like 14. I like the colors it comes in 15. Is functional for swimming Factor 1 -“Fashion” .796 .791 .108 .803 .039 .102 .754 .093 .762 .758 .043 .807 .810 .800 .106 Figure 6.7 Factor 2 -“Comfort” .061 .029 .782 .077 .729 .833 .059 .793 .123 .208 .756 .082 .055 .061 .798 6-12 12 Factor Scores Matrix Attribute 1. Attractive design 2. Stylish 3. Comfortable to wear 4. Fashionable 5. I feel good when I wear it 6. Is ideal for swimming 7. Looks like a designer label 8. Easy to swim in 9. In style 10. Great appearance 11. Comfortable to swim in 12. This is a desirable label 13. Gives me the look I like 14. I like the colors it comes in 15. Is functional for swimming Factor 1 -“Fashion” 0.145 0.146 -0.018 0.146 -0.028 -0.021 0.138 0.131 -0.021 0.146 -0.029 0.146 0.148 0.146 -0.019 Figure 6.8 Factor 2 -“Comfort” -0.022 -0.030 0.213 -0.017 0.201 0.227 -0.020 0.216 -0.003 0.021 0.208 -0.016 -0.024 -0.022 0.217 Sample calculation of factor scores: From the snake plot, the mean ratings of Aqualine on Attributes 1 through 15 are 2.15, 2.40, 3.48, …, 3.77. Multiply each of these mean ratings by the corresponding coefficient 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. 6-13 13 The AR Perceptual Map Comfort 2 Figure 6.9 Aqualine Gap 1 Islands Molokai Fashion -2 Splash Sunflare Gap 2 -2 2 6-14 14 Dissimilarity Matrix Aqualine Islands Sunflare Molokai Splash Aqualine X Islands 3 X Sunflare 9 8 X Molokai 5 3 5 X Figure 6.10 Splash 7 4 7 6 X 6-15 15 The OS Perceptual Map Aqualine Islands Splash Molokai Sunflare Figure 6.11 6-16 16 Comparing AR and OS Methods AR Methods Figure 6.12 OS Methods Input Required Brand ratings on specific attributes Overall similarity ratings Attributes must be pre-specified Respondent uses own judgment of similarity Analytic Procedures Commonly Used Factor analysis; multiple discriminant analysis Multidimensional scaling (MDS) Graphical Output Shows product positions on axes Shows product positions relative to each other Axes interpretable as underlying dimensions Axes obtained through follow-up analysis or must (factors) be interpreted by the researcher Where Used Situations where attributes are easily articulated or Situations where it may be difficult for the visualized respondent 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. 6-17 17 Failures 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?