Adaptive conjoint
Adaptive conjoint elicits preferences across product profiles with a configurable, adaptive-style exercise. You define attributes, how many questions to ask, how many profiles appear per question, and which estimation algorithm to use.
When to use it
Section titled “When to use it”Use adaptive conjoint when you need preference structure across multi-attribute products and want tighter control of interview length and adaptive options. Use standard Conjoint analysis when a generated fixed choice design fits better. Use MaxDiff for flat item importance.
How to add
Section titled “How to add”- Open Build → Form and select a page.
- Choose Add content → Adaptive conjoint (Pro).
- Define Product attributes and levels.
- Set max questions, profiles per question, algorithm, and optional price or none-of-these behavior.
Available in surveys, usability tests, and in-product surveys on the Pro plan.
Key settings
Section titled “Key settings”| Setting | Options | Notes |
|---|---|---|
| Max questions | 8, 12, 16, or 20 | Caps exercise length |
| Profiles per question | 2, 3, or 4 | Alternatives shown together |
| Algorithm | Hierarchical Bayes, Multinomial Logit, Random | Estimation approach |
| Show prices | On/off | Include price presentation when relevant |
| Allow none of these | On/off | Lets participants reject all profiles in a question |
| Product attributes | Attributes + levels | Same conceptual building blocks as conjoint |
Example
Section titled “Example”SaaS plans with attributes for seats, integrations, and support. Max questions: 12 · Profiles per question: 3 · Allow none of these: on · Algorithm: Hierarchical Bayes.
Results
Section titled “Results”Adaptive conjoint views include:
- Preference profile
- Customer segments
- Summary
Use segments carefully: confirm sample size before treating clusters as definitive personas.
Limitations
Section titled “Limitations”- Requires Pro.
- Longer max question settings increase fatigue.
- Algorithm choice should match how your team interprets utilities and segments.
- Start with fewer questions in pilots, then increase if estimates look unstable.
- Enable None of these when realistic markets include opting out.
- Keep attribute counts disciplined; adaptive flows still suffer when profiles feel artificial.
Common questions
How is adaptive conjoint different from conjoint analysis?
Adaptive conjoint adjusts the questioning flow based on settings such as max questions, profiles per question, and algorithm. Standard conjoint analysis focuses on generated fixed combinations from attributes and levels.
Which algorithms are available?
Hierarchical Bayes, Multinomial Logit, and Random. Choose based on your analysis preference and study design guidance.
What appears in results?
Preference profile, customer segments, and a summary view help you interpret adaptive conjoint outcomes.