Sensory Informed Design: An effective clustering of incomplete block consumer data
•
3 likes•979 views
Consumer research has advanced its business relevance through segmenting consumer populations into clusters based upon liking. Products designed to meet the expectations and desires of specific niche markets have demonstrated commercial success. The studies that are typically designed to reveal liking segments require a relatively large number of products and a large sample of consumers in a complete block design. A study of 12 Cabernet Sauvignon wines was conducted using over 600 consumers recruited and tested for liking of 3 of the 12 wines in a BIB design. The data were subsequently analyzed for liking clusters with missing data replaced with the consumer’s individual mean. Four liking clusters successfully demonstrated different sensory liking profiles. The method was not robust. Consequently, a research program was initiated to develop a systematic approach to building designs using sensory information to ensure contrast. The Sensory Informed Design (SID) approach was applied to a 12-present-6 study of white breads. All breads were profiled using calibrated descriptive analysis. The results of the DA were used to construct a balanced experimental design (12:6) that included two smaller-sized SIDs (12:3 and 12:4) nested within the experiment to evaluate the efficiency and stability. Consumer data (n=400) were collected and missing data were imputed as part of a novel EM approach for mixture model-based clustering; the one latent factor model gave a six-cluster solution. In 2012, a study of whole grain breads was conducted with 570 consumers using an improved SID of 16:6, with nested designs of 16:3 and 16:4. The nested designs demonstrated stable clusters, provided internal validation and supported the results of previous work. The application of SID, EM imputation and model-based cluster analysis can dramatically reduce the resources required to conduct large category appraisals and deliver effective consumer clusters.
Recommended
Recommended
More Related Content
Viewers also liked
Viewers also liked (10)
Similar to Sensory Informed Design: An effective clustering of incomplete block consumer data
Similar to Sensory Informed Design: An effective clustering of incomplete block consumer data (20)
More from Compusense Inc.
More from Compusense Inc. (11)
Recently uploaded
Recently uploaded (20)
Sensory Informed Design: An effective clustering of incomplete block consumer data
- 1. Sensory informed design: An effective clustering of incomplete block consumer data Chris Findlay, Compusense Inc. Ryan Brown, Department of Mathematics and Statistics, University of Guelph Paul McNicholas, Department of Mathematics and Statistics, University of Guelph John Castura, Compusense Inc.
- 2. Sensory informed design • Pangborn 2013 • Consumer segmentation is essential to the understand of liking • Consumer-driven product development works • Large consumer tests are expensive • Large consumer tests take a lot of time and resources Consumer Category Tests
- 3. Sensory informed design • Pangborn 2013 • The Risks – Consumer Fatigue – Carry-over effect – Boredom – Consumers behaving like experts – Resource restrictions • The Remedies – Testing at a single event – Incomplete Block Designs • The Challenges – Missing data – Validation Considerations about Large Studies
- 4. Sensory informed design • Pangborn 2013 The Effect of Order and Day on Consumer Liking 12 White Wines, 115 Consumers, CBD 12:12 over 3 Days 0 10 20 30 40 50 60 70 1 2 3 All Day 1 Day 2 Day 3 Combined Positions 1st 2nd 3rd 4th
- 5. Sensory informed design • Pangborn 2013 The Statistical Challenge • A valid approach to segmentation of consumer BIB data • Using a combination of sensory best practice, experimental design and advanced statistical analysis We will explore this approach using three different studies: Cabernet Sauvignon Study White Bread Study Whole Grain Bread study Sensory Informed Design Method Development
- 6. Cabernet Sauvignon Study • A study of 12 Cabernet Sauvignon wines was conducted using over 600 recruited consumers and tested for liking • Consumers sampled 3 of the 12 wines in a BIB design • Data was analyzed for liking clusters with missing data replaced with consumer’s individual mean • Four liking clusters successfully demonstrated different sensory liking profiles Sensory informed design • Pangborn 2013
- 7. Sensory informed design • Pangborn 2013 3 4 5 6 7 8 9 W1 W2 W3 W4 W5 W6 W7 W8 W9 W10 W11 W12 Cabernet Sauvignon Mean Liking
- 8. Sensory informed design • Pangborn 2013 3 4 5 6 7 8 9 W1 W2 W3 W4 W5 W6 W7 W8 W9 W10 W11 W12 3 4 5 6 7 8 9 W1 W2 W3 W4 W5 W6 W7 W8 W9 W10 W11 W12 3 4 5 6 7 8 9 W1 W2 W3 W4 W5 W6 W7 W8 W9 W10 W11 W12 3 4 5 6 7 8 9 W1 W2 W3 W4 W5 W6 W7 W8 W9 W10 W11 W12 Cluster 1 – 28% Cluster 2 – 23% Cluster 3 –32% Cluster 4 – 17%
- 9. Cabernet Sauvignon Study: Results • Findings demonstrated that although the method was not robust, the approach gave useful and actionable results • A research program was initiated to develop a systematic approach to building designs using sensory information to ensure contrast Sensory informed design • Pangborn 2013
- 10. Sensory informed design • Pangborn 2013 • To state a true preference a consumer must be able to perceive a real difference • Otherwise it’s just a guess • Consequently we must present the consumer with truly different samples, by design. Sensory Design Is there a real Preference?
- 11. Sensory informed design • Pangborn 2013 • Let’s consider a sensory space • Can we find logical contrasts to test Factor Scores plot: dimension 1 versus 2 -2.58 2.58 -2.58 2.58 Tobacco Smoke Asparagus Coffee Floral Vanilla Green Woody Pepper EucalyptusFruity Raisin Leather Sweet Sour Bitter Astringent W1 W2 W3 W4 W5 W6 W7 W8 W9 W10 W11 W12 Factor Scores plot: dimension 3 versus 4 -2.58 2.58 -2.58 2.58 Tobacco Smoke Asparagus CoffeeFloral Vanilla Green WoodyPepper Eucalyptus Fruity Raisin Leather Sweet Sour Bitter Astringent W1 W2W3 W4 W5 W6 W7 W8 W9 W10 W11 W12
- 12. Sensory informed design • Pangborn 2013 • Quadrangles and Triangles Factor Scores plot: dimension 1 versus 2 -2.58 2.58 -2.58 2.58 Tobacco Smoke Asparagus Coffee Floral Vanilla Green Woody Pepper EucalyptusFruity Raisin Leather Sweet Sour Bitter Astringent W1 W2 W3 W4 W5 W6 W7 W8 W9 W10 W11 W12 Factor Scores plot: dimension 3 versus 4 -2.58 2.58 -2.58 2.58 Tobacco Smoke Asparagus CoffeeFloral Vanilla Green WoodyPepper Eucalyptus Fruity Raisin Leather Sweet Sour Bitter Astringent W1 W2W3 W4 W5 W6 W7 W8 W9 W10 W11 W12
- 13. White Bread Study • All breads were profiled using calibrated descriptive analysis and using a trained panel • The Sensory Informed Design (SID) was used to construct a balanced incomplete block design (12:6) • Two smaller SIDs (12:3 and 12:4) were nested within the experiment to evaluate efficiency and stability. • 400 category consumers • 200 observations per product Sensory informed design • Pangborn 2013
- 14. Sensory informed design • Pangborn 2013 White Bread Mean Liking 3 4 5 6 7 8 9 B1 B2 B3 B4 B5 B6 B7 B8 B9 B10 B11 B12
- 15. Sensory informed design • Pangborn 2013 Cluster 1 – 22% Cluster 2 – 36% Cluster 3 –20% Cluster 4 – 22% 3 4 5 6 7 8 9 B1 B2 B3 B4 B5 B6 B7 B8 B9 B10 B11 B12 3 4 5 6 7 8 9 B1 B2 B3 B4 B5 B6 B7 B8 B9 B10 B11 B12 3 4 5 6 7 8 9 B1 B2 B3 B4 B5 B6 B7 B8 B9 B10 B11 B12 3 4 5 6 7 8 9 B1 B2 B3 B4 B5 B6 B7 B8 B9 B10 B11 B12
- 16. White Bread Study: Results • Consumer data (n=400) was collected and missing data was imputed as part of a novel EM approach for mixture model-based clustering • The scatter plots below demonstrate the stability of the clusters across all three partial present blocks Sensory informed design • Pangborn 2013
- 17. Whole Grain Bread Study In a 2012 study of whole grain breads, 570 consumers 16 samples using an improved SID of 16:6, with nested designs of 16:3 and 16:4 Sensory informed design • Pangborn 2013
- 18. Sensory informed design • Pangborn 2013 GPA of 16 Whole Grain Breads 55 Sensory Attributes -1.10 1.10 -0.50 0.50 Variance Accounted for: DIM1 – 66% DIM2 – 9% DIM3 – 7% DIM4 – 4%
- 19. Sensory informed design • Pangborn 2013 3 4 5 6 7 8 9 Overall Liking All 570 Category Consumers 19 7.1 6.8 6.6 6.5 6.4 6.2 5.9 5.9 5.8 5.8 5.7 5.5 5.5 5.2 5.0 7.3 LOVE IT! HATE IT!
- 20. Sensory informed design • Pangborn 2013 2 3 4 5 6 7 8 9 Cluster 1 Overall Liking 25.8 % 147 Consumers 20 7.4 6.7 6.7 6.6 6.5 6.1 5.8 6.0 4.9 5.6 4.9 4.4 4.5 4.4 3.2 7.8 LOVE IT! HATE IT!
- 21. Sensory informed design • Pangborn 2013 3 4 5 6 7 8 9 Cluster 2 Overall Liking 45.3 % 258 Consumers 21 7.9 7.8 7.7 7.5 7.3 7.0 6.6 6.9 6.3 6.5 6.0 5.9 5.9 5.7 5.6 8.0 LOVE IT! HATE IT!
- 22. Sensory informed design • Pangborn 2013 3 4 5 6 7 8 9 Cluster 3 Overall Liking 28.9 % 165 Consumers 22 6.9 6.2 6.1 5.9 5.7 5.5 5.3 5.5 4.6 4.9 4.6 4.5 4.5 4.5 3.7 7.6 LOVE IT! HATE IT!
- 23. Sensory informed design • Pangborn 2013 Demonstrate stable clusters The cluster membership remains consistent Provide internal validation Shows the same outcome independently Nested Designs and Validation
- 24. Sensory informed design • Pangborn 2013 1. Calibrated DA of all products defines the sensory space, followed by the creation of a nested experimental design based upon sensory contrasts using 3 and 4 samples for each consumer data set 2. Imputation of the missing data using an advanced EM (Expectation Maximization) algorithm (http://arxiv.org/abs/1302.6625) 3. Model-based cluster analysis to ensure a stable clustering solution Key elements of Sensory Informed Design
- 25. Sensory informed design • Pangborn 2013 • Improves the efficiency of large studies • Can be used for any product category • Improves the quality of data collected • Delivers actionable consumer clusters • Saves resources Conclusions about Sensory Informed Design
- 26. Sensory informed design • Pangborn 2013 For more information, please contact Chris Findlay, PhD Chairman +1 800 367 6666 (North America) +1 519 836 9993 (International) cfindlay@compusense.com