How to Use Conjoint Analysis In the Innovation Process?
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 SurveyAnalytics, responded to each one.
At SurveyAnalytics we offer a robust yet easy to use Discrete Choice Conjoint Analysis tool. Guidelines are provided to ensure data is concise and accurate. We also provide a market segmentation tool, which offers you an opportunity to test new product ideas against your current data to help predict possible market share.
DS . It's important in the lead in that you let the respondent know that YOU are taking the survey seriously and that you would appreciate if they do also. This is less of a problem if you're using your own databases. You should also try to screen out responses that are obviously completed just to finish, such as never varying their response. SA . I agree with Dorian. Respondents always appreciate an introduction that is upfront with your intentions. Be honest with how long it would possibly take and provide an incentive that appeals to your targeted sample. In my experience of working with internal databases, you will become familiar with those who are not truthful or do not take your surveys seriously and can remove them from future surveys.
This depends on your target market. The larger your target market, the larger your sample should be for statistically significant data. The general rule of thumb for Conjoint Analysis is usually a minimum of 200-300 completed surveys. This, however you can go down to 100 completed surveys if your target market is relatively small.
It would depend if the feature is something you may want to add or not.
For example, if you wanted Trail Mix with/without Crackers you would set up the following:
Features: Crackers >> Level: Yes, No
It has been investigated in other research and will be tested again further.
Can be used similarly. E.g. Instead of price it may be price/mo., etc. You must identify attributes and levels similar to a product.
SA . A fun example is a hair salon. What kinds of services will you offer to your clients and at what price do you think they would pay for it?As Dorian said you must identify attributes and levels similar to a product.
The case study survey that was used during the presentation took respondents on average 15 minutes to complete.
Yes, Conjoint Analysis can be used in any industry that is interested in doing a trade-off analysis of some type. Whether it is on a medication a pharmaceutical company is trying to develop or a new kayak model that would appeal to families with young children, Conjoint Analysis can be used to provide guidance in those industries.
The minimum is 2 levels per feature/attribute. The standard is to stick to no more than 3-4 levels per feature/attribute. Every once in a while going up to 5 may be needed depending on the feature needed to be test.
You'll want to keep the number of attributes and levels reasonably low.
The fewer the respondents being surveyed the fewer attributes and levels should be used. At this point in your research you should have highly defined features and levels that would fit your targeted sample size.
From a technical standpoint, the system does NOT impose any limitations. You can have unlimited attributes and unlimited levels within each attribute.
However, from a practical standpoint, it is unreasonable to have more than 4-6 attributes, and about 3-4 levels per attribute. Our suggestion would be to keep the number of attributes to under 5 and try and seek about 3 levels for each attribute.
Our experience has shown that there is a precipitous dropout rate after about 15 tasks. Unless there is a strong personal incentive for the end-users to complete the survey, we would suggest keeping the number of tasks to fewer than 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.
On the lower side, we would suggest that 5-8 tasks be the minimum for a conjoint model with 3 attributes. The more attributes you have, the more number of tasks users has to fill out.
SA . Yes, that is correct.
This is true, but this is part of a conjoint analysis to understand what your customers deem which attributes and levels are the worst. I don't think you want to limit options for a high price and low attribute levels.
We have built intelligence into our conjoint tool such as the prohibited pairs tool to ensure certain combinations that are not possible will ever show up. We must be careful in using this tool because the idea is not to limit the profiles based on what the client will not do, but to find out what resonates higher with your audience. We also provide a concept simulator that will calculate the number of times an attribute will be shown given the approximate number of people who will complete the survey.
No you do not have access to this tool. To access the Discrete Choice Conjoint Analysis tool you must upgrade to SurveyAnalytics.
This depends on what you are interested in retrieving data for. If you are looking for data that mimics the purchase process then the Choice Based (Discrete Choice) Conjoint Analysis is the better bet. Adaptive Conjoint Analysis (ACA) is a computer-administered, interactive conjoint method designed for situations in which the number of attributes exceeds what can reasonably done with Choice Based Conjoint Analysis.
SurveyAnalytics specializes in Discrete Choice Conjoint Analysis.
It's important to focus on the most important attributes that really drive decisions. You'll probably want to do preliminary research such as interviews, focus groups or short quantitative to narrow it down.
Market Share: % of Profile 1 / All profiles in simulator based on % on relative importance and #of responses seen. The market simulator uses aggregate utility values to project the probability of choice and hence the market share