MaxDiff and Conjoint (CBC) Design and Sample Size Calculator
Plan MaxDiff and choice-based conjoint studies: sets per respondent, item exposures, design size and minimum sample using Orme's rule of thumb.
inquiry@globalvoxpopuli.com · globalvoxpopuli.com
How to use the MaxDiff and conjoint calculator
- Choose MaxDiff (best–worst scaling) or choice-based conjoint (CBC).
- For MaxDiff, enter the number of items, items shown per set and sets per respondent. For conjoint, list the number of levels for each attribute, the tasks per respondent and concepts per task.
- Enter the number of subgroups you will report separately and the sample you plan for each.
MaxDiff design rules of thumb
- Show 3 to 5 items per set. Four is the most common.
- Show each item 3 or more times per respondent if you want individual-level scores (hierarchical Bayes).
- The number of sets needed is roughly 3 × items ÷ items per set. Twenty items in sets of four needs about 15 sets.
- Plan for 200 or more respondents per subgroup you will compare; 300+ gives more stable results.
Conjoint sample size: Orme's rule of thumb
n ≥ 500 × c ÷ (t × a)
Here c is the largest number of levels in any attribute, t the number of tasks and a the number of concepts per task. With a 6-level attribute, 10 tasks and 3 concepts, the minimum is 100 per subgroup. In practice, most commercial conjoint studies use at least 300 per subgroup so that segment-level simulations are stable.
Worked example: a three-country pricing study
A durable goods brand tests 6 attributes with 6, 5, 4, 3, 3 and 2 levels, using 10 tasks of 3 concepts, in three countries.
- Orme minimum: 500 × 6 ÷ 30 = 100 per country
- Recommended: 300 per country, 900 in total
- The design has 6 × 5 × 4 × 3 × 3 × 2 = 2,160 possible products, and each respondent sees 30.
Press Example to load it.
Frequently asked questions
How many respondents do I need for MaxDiff?
Usually 200–300 per subgroup you want to compare, and more if items are shown fewer than 3 times each per respondent.
How many attributes can a conjoint have?
Full-profile CBC works best with up to 6 or 7 attributes. Beyond that, use partial-profile or adaptive designs.
How long does a MaxDiff or conjoint exercise take?
MaxDiff sets take roughly 15–20 seconds each; CBC tasks 20–40 seconds depending on concepts and attributes. The survey length calculator adds them to the rest of the questionnaire.
Need the respondents, not just the numbers?
Global Vox Populi runs quantitative and qualitative fieldwork in 170+ countries through its own proprietary panels of consumers, B2B and IT decision-makers, physicians, nurses, patients, caregivers and payers. ISO 9001, ISO 20252 and ISO/IEC 27001 certified and HIPAA compliant, following ESOMAR and Insights Association guidelines.
Check feasibilityLink to this tool
Free to use in proposals, courses and articles. If it helped, a link back is appreciated:
<a href="https://globalvoxpopuli.com/tools/maxdiff-conjoint-sample-size-calculator/">MaxDiff & Conjoint Design Calculator</a> by Global Vox Populi