Conjoint Analysis is a powerful and often under-utilized marketing research tool that can provide powerful insight into how your customers actually think. The resulting information can be used to prioritize features, develop pricing strategies, and estimate market share… all before you develop your product or spend valuable marketing dollars.
Participants posted the following questions and both presenters, Dorian Simpson of Planning Innovations and Esther LaVielle of Survey Analytics, responded to each one.
1) I Do straightforward quant work like cross tabs, but I wonder if I can really do conjoint - does it take super high skill to set up and analyze the conjoint, or can a relative newcomer do a decent job. Should I try to work with a mentor?
A: There are a lot of resources and white papers available to learn about conjoint analysis. When working with SurveyAnalytics, you will have a dedicated account manager who walk you through your projects, offer guidance and tips to get statistically significant data. You may also work directly with consultants such as Dorian Simpson from Planning Innovations who can train and help your understanding of conjoint analysis.
Q: Can you spend a minute or so explaining how the conjoint questions get embedded ina survey? does it integrate with outside survey tools? What is the form of the data output? Excel? SPSS?
A: The conjoint question template can we used in conjunction with other question types in SurveyAnalytics so this does not need to be embedded into another survey tool to work.
Q: How do we choose between a full factorial design (all combinations, costly but highly accurate approach) versus a partial factorial design (selected combinations of attribute levels, cost-effective, relatively less accurate) in the conjoint analysis studies?
A: We do partial. We generate a orthogonal set of profiles which you then chose how many the respondent sees. Full factorial is something we add later, but as you say, it is costly.
Q: How do you determine what "task count" you assign to a given study? Presumeably there is a trade-off here between overburdening the respondent and getting low quality results? Are there rules of thumb you use?
A: Our experience has shown that there is a precipitious drop-out rate after about 15 tasks. Unless there is a strong personal incentive for the end-users to complete the survey, we would suggest to keep the number of tasks to under 15 especially in cases where users are volunteering to take surveys. Please keep in mind that conjoint product selection is a little more involved than simply "answering a survey question" -- users have to comprehend each of the attributes/concepts and then make a choice. This is a lot more involved than say choosing "Male/Female" on a gender question.
On the lower side, we would suggest that 6-8 tasks be the minium for a conjoint model with 3 attributes. The more attributes you have, the more number of tasks users have to fill out.
It is obvious that it's a balancing act between the number of tasks, concepts per task and the total number of attributes/levels than need to be displayed.
Two factors determine the overall utility:
- Concepts Per Task
- Total # of Tasks
The system provides the "Concept Simulator" - with the concept simulator you can see the TOTAL number of times a particular level will be displayed (given the total number of respondents)
Q: Is it important to describe attributes or is it better to let less knowledgeable respondents use their best guess what it means?
A: This depends on many factors. You may want to review how homogeneous your respondent list is, number of times you've surveyed this particular group, and also if the features and levels you are presenting are familiar to your respondents or not.
With SurveyAnalytics, you can add tips, definitions, instructions and visuals for respondents to make a better trade-off decision.
Q: Are slides available?
A: Yes. Click below to review slides from both presenters
About the Presenters:
Dorian Simpson founded Planning Innovations in 2002 to help technology-driven companies launch successful products and services through focused innovation management and planning. He has significant experience in both engineering and marketing to help build the bridge between these two critical innovation functions.
Esther LaVielle is a Senior Account Manager at QuestionPro and Survey Analytics, which was started in 2002 in Seattle and is now one of the fastest growing private companies in the US. Prior to her adventure at QuestionPro she spent 3 years as a Qualitative Project Manager at the Gilmore Research Group.