What is the optimal number of diagnostics in a Claims Test or Product Variant Selector?
We consider the optimal number of diagnostics from 3 to 7 questions to achieve an efficient and adequate experiment design for a Claims Test or Product Variant Selector.
For instance, suppose a claims test experiment with n claims and 7 diagnostics. Each respondent will answer diagnostics for 5 claims by default, so this means each respondent is going to answer 35 diagnostic questions (5 claims * 7 diagnostics). It would be fatiguing for respondents to answer more than 35 diagnostic questions and it could also affect the results of the analysis. On the other hand, a design with less than 3 diagnostics could limit the comprehension of preferences and insights for each claim.
Therefore, we suggest including between 3 and 7 diagnostic questions, making it easier for respondents rate statements for standard questions in the survey.