Sweet Tooth Bakery Cake preference conjoint analysis Michael Bystry, Stacy Comrie, Shannon Goyda, Prochi Jain, Amy Kasten
Background/Problem Statement <ul><li>Sweet Tooth Bakery wants to create a new cake offering for its customers </li></ul><u...
Background/Problem Statement <ul><li>Cake components: </li></ul><ul><ul><li>Cake Flavor </li></ul></ul><ul><ul><ul><li>Whi...
Design <ul><li>Full-profile conjoint </li></ul><ul><ul><li>Minimum design size=9 profiles </li></ul></ul><ul><ul><li>Full ...
Analysis and Results <ul><li>44 usable completes </li></ul><ul><li>Validation procedure </li></ul><ul><ul><li>Correlations...
Analysis and Results <ul><li>Importance and Part-worths </li></ul><ul><ul><li>Average, maximum, and minimum part-worths ca...
Analysis and Results
<ul><li>Share Simulator Usage </li></ul><ul><li>Base State </li></ul><ul><li>Updated with new selections </li></ul>Analysi...
Analysis and Results <ul><li>Customer clusters </li></ul><ul><ul><li>Segmentation based on importance scores as derived fr...
Analysis and Results <ul><li>Segments of cake buyers </li></ul><ul><ul><li>1.  Frosting/filling segment  (38%) </li></ul><...
Analysis and Results
Limitations <ul><li>Due to the nature of conjoint analysis </li></ul><ul><ul><li>No information about purchase intent </li...
Limitations <ul><li>Due to limited time and resources </li></ul><ul><ul><li>Respondents drawn from convenience sample </li...
Questions/Comments?
Upcoming SlideShare
Loading in …5
×

Conjoint class project

1,293 views

Published on

This is conjoint class project to assist a bakery in optimizing its product offerings. It was completed as part of a Conjoint/Discrete Choice for the MMR program at UGA. Any questions can be sent to michaelbystry@yahoo.com.

0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
1,293
On SlideShare
0
From Embeds
0
Number of Embeds
8
Actions
Shares
0
Downloads
40
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Conjoint class project

  1. 1. Sweet Tooth Bakery Cake preference conjoint analysis Michael Bystry, Stacy Comrie, Shannon Goyda, Prochi Jain, Amy Kasten
  2. 2. Background/Problem Statement <ul><li>Sweet Tooth Bakery wants to create a new cake offering for its customers </li></ul><ul><li>The cake will be 9”x13” and sell for $17.99 </li></ul><ul><li>In order to build the most appealing cake, the bakery needs to understand consumer preference for different cake components </li></ul>
  3. 3. Background/Problem Statement <ul><li>Cake components: </li></ul><ul><ul><li>Cake Flavor </li></ul></ul><ul><ul><ul><li>White, Chocolate, Yellow, Marble </li></ul></ul></ul><ul><ul><li>Frosting/Filling Flavor </li></ul></ul><ul><ul><ul><li>Chocolate/Raspberry, Vanilla/Raspberry, Vanilla/Lemon, Cream Cheese/Raspberry, Cream Cheese/Lemon </li></ul></ul></ul><ul><ul><li>Number of Layers </li></ul></ul><ul><ul><ul><li>Two or three </li></ul></ul></ul>
  4. 4. Design <ul><li>Full-profile conjoint </li></ul><ul><ul><li>Minimum design size=9 profiles </li></ul></ul><ul><ul><li>Full factorial design=40 profiles </li></ul></ul><ul><ul><li>Selected Design=20 profiles </li></ul></ul><ul><ul><ul><li>D-efficiency=99.5% with only one violation </li></ul></ul></ul><ul><ul><ul><li>No canonical correlations >.316 </li></ul></ul></ul><ul><ul><li>Two hold-out tasks added to the design </li></ul></ul><ul><ul><ul><li>Final design size=22 profiles </li></ul></ul></ul><ul><ul><ul><ul><li>D-efficiency=98.7% </li></ul></ul></ul></ul><ul><ul><ul><ul><li>No canonical correlations >.316 </li></ul></ul></ul></ul>
  5. 5. Analysis and Results <ul><li>44 usable completes </li></ul><ul><li>Validation procedure </li></ul><ul><ul><li>Correlations between 20 test profiles and two holdout profiles were examined </li></ul></ul><ul><ul><ul><li>Seven respondents removed </li></ul></ul></ul><ul><ul><ul><ul><li>Responses to holdout tasks not consistent with responses to test profiles </li></ul></ul></ul></ul><ul><ul><ul><li>37 remaining respondents used for analysis of importance scores and market share </li></ul></ul></ul>
  6. 6. Analysis and Results <ul><li>Importance and Part-worths </li></ul><ul><ul><li>Average, maximum, and minimum part-worths calculated for each attribute </li></ul></ul><ul><ul><li>Importance scores calculated for each attribute </li></ul></ul>Cake Component Highest Average Part-Worths Lowest Average Part-Worths Importance Scores Cake White (5.2) Marble (-4.69) 37% Frosting/filling Chocolate/Raspberry (11.12) Cream cheese/lemon (-14.23) 55% Layers Two (0.98) Three (-.098) 8%
  7. 7. Analysis and Results
  8. 8. <ul><li>Share Simulator Usage </li></ul><ul><li>Base State </li></ul><ul><li>Updated with new selections </li></ul>Analysis and Results
  9. 9. Analysis and Results <ul><li>Customer clusters </li></ul><ul><ul><li>Segmentation based on importance scores as derived from individuals’ utility functions </li></ul></ul><ul><ul><li>Two-, three-, and four-cluster solutions examined </li></ul></ul><ul><ul><ul><li>Three-cluster solution gave best results </li></ul></ul></ul>
  10. 10. Analysis and Results <ul><li>Segments of cake buyers </li></ul><ul><ul><li>1. Frosting/filling segment (38%) </li></ul></ul><ul><ul><ul><li>High importance given to frosting/filling attribute </li></ul></ul></ul><ul><ul><li>2. Cake flavor segment (19%) </li></ul></ul><ul><ul><ul><li>High importance given to cake flavor attribute </li></ul></ul></ul><ul><ul><li>3. Holistic segment (43%) </li></ul></ul><ul><ul><ul><li>Higher than average importance give to layers </li></ul></ul></ul><ul><ul><ul><li>Relative equality given to cake flavor and frosting/filling </li></ul></ul></ul>
  11. 11. Analysis and Results
  12. 12. Limitations <ul><li>Due to the nature of conjoint analysis </li></ul><ul><ul><li>No information about purchase intent </li></ul></ul><ul><ul><ul><li>Only preference information </li></ul></ul></ul><ul><ul><li>All possible options not included in design </li></ul></ul><ul><ul><ul><li>Some cake, filling, and frosting options removed </li></ul></ul></ul><ul><ul><li>No information about possible interactions </li></ul></ul><ul><ul><ul><li>Frosting/filling may interact with cake flavor </li></ul></ul></ul><ul><ul><li>The model doesn’t capture pricing information </li></ul></ul><ul><ul><ul><li>Conjoint not well suited to capturing price </li></ul></ul></ul>
  13. 13. Limitations <ul><li>Due to limited time and resources </li></ul><ul><ul><li>Respondents drawn from convenience sample </li></ul></ul><ul><ul><li>Sample size is too small for statistically-meaningful results </li></ul></ul><ul><ul><ul><li>Results can not be projected onto the general population </li></ul></ul></ul><ul><ul><li>Sample size is too small to allow for use of holdout sample for validation </li></ul></ul><ul><ul><ul><li>Holdout tasks were used instead </li></ul></ul></ul>
  14. 14. Questions/Comments?

×