MaxDiff, conjoint and pricing modules, programmed so the design survives the build
Choice designs fail quietly. A version that is never assigned, a task that shows the wrong items or an export that drops what was shown does not break the survey; it breaks the model, and usually only the analyst finds out. Wisfield programs choice and pricing exercises in Decipher with the design treated as part of the specification, and tests it as such.
What is covered
- MaxDiff (best-worst scaling), including anchored designs
- Choice-based conjoint from a supplied design, with attributes, levels and prohibitions as specified
- Gabor-Granger price ladders, with start points and stop rules set out in the specification
- Monadic and sequential monadic concept tests with balanced rotation
How the design is protected
- The design is checked against the specification before build: versions, tasks and items per task
- Version assignment is balanced and recorded for every respondent
- Each task stores what was shown as well as what was chosen, so the data can be modelled without guesswork
- Test data is run through the exercise to confirm counts per item and per version before launch
- The independent audit covers the exercise screen by screen, on desktop and mobile
Timeline and scope
Choice and pricing studies are scoped individually, because the logic decides the time, not the length. The written scope states the build, audit and test days. The sample written scope (PDF) shows one for a physician study combining a monadic concept test and a MaxDiff.
The model is only as good as the data the survey records. That is checked before launch, not after.
Written Scoping
Send the questionnaire. Receive a written scope.
A written scope and a fixed quote, with the main risks flagged, within two business days.