Class project ds_604_revise
- 1. Conjoint Analysis Ryan Louie, I-Hua Lin, Letty Kwok
- 2. Conjoint Analysis Used when marketing a new product/service, pricing purposes, forecasting of the product/service Also known as Discrete Choice, Choice Modeling, Hierarchical Choice, Tradeoff Matrices The goal of Conjoint Analysis is to assign specific values to the range of options buyers have when considering making a purchasing decision
- 3. Model Advantages Able to use physical objects Measures preferences at the individual level Estimates psychological tradeoffs that consumers make when evaluating several attributes together
- 4. Model Disadvantages Only a limited set of features can be used because the number of combinations increase quickly as more features are added Information gathering stage can be complex Respondents are unable to articulate attitudes toward new categories Time lag
- 5. Defining a Product or Service Define each attribute’s (i) importance (ii) most desired level for each consumer
- 6. Study Design 1. Determine Relevant Attributes and their levels 2. Determine Product Presentation: Content and Form 3. Determine Respondent - Researcher Interaction format 4. Decide on Response Type , rating, ranking, or choice 5. Determine Criterion for judging, liking, purchasing or willing to pay 6. Decide on Data Analysis Technique
- 7. from “conjoint and discrete designs for pricing research” by Vinay Kanetkar
- 8. Analysis the Outputs 1. Aggregate analysis 2. Segmentation analysis 3. Scenario simulations
- 9. Aggregate Analysis To give the analyst an overall feeling for: (i) which attributes are generally important (ii) what is the most desired level of each attribute
- 10. Example Monthly Avg. Access Value Phone Avg. Value Contract Avg. Charge Price Term Value $12.95 0.000 Free 0.000 1 year 0.000 $24.95 -0.921 $29.95 -1.174 2 years -1.181 $44.95 -3.406 $69.95 -3.138 3 years -1.864 $79.95 -5.677 $119.95 -3.142 from “conjoint and discrete designs for pricing research” by Vinay Kanetkar
- 11. Range Monthly access charge = 0-(-5.677) = 5.677 Phone price = 0-(-3.142) = 3.142 Contract term =0-(-1.864) = 1.864
- 12. Monthly access charges =5.677/(5.677 + 3.142 + 1.864) = 53.18% Phone price =3.142/(5.677 + 3.142 + 1.864) = 29.4% Contract term =1.864/(5.677 + 3.142 + 1.864) = 17.4%
- 13. Segmentation Analysis Strategy to analysis on customers segment Cluster analysis can be used to produce “benefit segments” Look at predefined group of customers based on some prior knowledge about them
- 14. Example:
- 15. Example:
- 16. Scenario Simulations Predicting market shares or unit sales in various scenarios
- 18. Example: from “conjoint and discrete designs for pricing research” by Vinay Kanetkar
- 20. Accuracy There are three methods to verify accuracy Common-sense test, or face validity Holdout prediction Actual vs. predicted market share