Monadic vs Sequential Monadic, and Which Research Design to Use

A stimulus is anything a respondent evaluates: a concept, a pack, an ad, a name, a logo or a price. The cell designs differ in how many stimuli one respondent sees and in what order.

Monadic vs sequential monadic. Choose pure monadic when seeing one idea would change how people read the next. Taste, fragrance, price and final ad validation are the usual cases. Choose sequential monadic when you have 2 to 4 ideas, each is quick to read, and the question is which is stronger.

Rotate the order in a sequential design. With 3 or more stimuli, use a Latin square so each idea appears in each position equally often. Then check the position effect on your own data: compare each idea's score when it came first with its score when it came last. If first-position scores run clearly higher, report first-position reads only, or rerun as pure monadic.

More than 4 stimuli? Do not push them all through one sequential cell. Run a MaxDiff or a screening round first, then carry the top 2 to 4 into a monadic test.

MaxDiff shows a few items at a time and asks which is best and which is worst. Over several screens you get a ratio-scaled order for 8 to 30 items. Rating grids let every item score 4 or 5. MaxDiff cannot. See the MaxDiff guide for set design and analysis.

Kano asks two questions per feature: how you would feel if it were present, and how you would feel if it were absent. The answers classify each feature as must-be, performance, attractive, indifferent or reverse. Keep the list to about 15 features. Read the Kano guide for the evaluation table.

The two answer different questions. MaxDiff tells you how much people value a feature relative to the others. Kano tells you what happens if it is missing. A feature can rank low on MaxDiff and still be a must-be that annoys people when absent.

TURF finds the set of items that reaches the most people. It needs a per-item answer, such as would buy or would use. The best set is not the top individual items, because the top items often appeal to the same people. Simple ranking works for 3 to 7 items. Beyond that, respondents tire and the bottom of the list is noise.

Pick the price design by the question you need answered. Our pricing research guide owns the method comparison; this is the short version.

• Van Westendorp asks four open price questions and returns an acceptable range and an optimal point. Use it for a new product with no price anchor. Details are in the Van Westendorp guide.
• Gabor-Granger asks buy likelihood on a 5-point scale at each rung of a price ladder, and gives a demand curve and a revenue-optimal price. Use it when you have a short list of candidate prices. See the Gabor-Granger guide.
• Monadic price cells show each group one price. This avoids anchoring, because no one sees a second price. It costs a cell per price.

For Van Westendorp and Gabor-Granger, treat our platform as the questionnaire builder: it supports the question design and routing, and you compute the curves in your own analysis.

Suppose you test 3 concepts and want 200 reads per concept. The numbers below are illustrative, priced at our list rates (pricing page), and the 200-per-concept target is a planning assumption, not a rule.

Sequential monadic costs less because each person is counted three times. The trade is order effects and a longer survey. If you rotate fully, only about a third of your reads for each concept are first-position reads, so a clean first-position read at 200 per concept needs 600 respondents anyway.

Precision. At 200 respondents and a share near 50%, the margin of error on one score is about 6.9 points. To call two independent cells different when both sit near 40%, the gap needs to be about 9.6 points. At 400 per cell it is about 6.8 points (both from the standard formula for two proportions at 95% confidence). Within-person comparisons in a sequential design usually detect smaller gaps, because each person is their own comparison. How much smaller depends on how closely the two answers move together. Read our significance testing guide before declaring a winner.

Keep the survey to 8 to 12 minutes on a smartphone. Each extra stimulus adds a full battery of questions, so the length limit often decides between sequential and pure monadic before the budget does.

If your job is a concept, pack, ad or price comparison among consumers in India, our survey builder covers the designs above: single and pure monadic, sequential and proto-monadic, reverse sequential, paired comparison, side-by-side, MaxDiff, Kano, TURF, ranking, Van Westendorp, Gabor-Granger and monadic price cells. It rotates and randomises stimuli for you. It does not build choice-based conjoint.

Pro is sold in bundles of 400 to 1,000 responses and costs 20% less per response than Plus. Enterprise plans come with a dedicated research team and are priced on request.

Respondents come from the SuperJ app, which has a 25M+ consumer footprint. The survey-eligible panel is a subset, and what you can reach depends on your targeting. Members complete Aadhaar-based KYC once, then take part through a zero-knowledge proof. Surveys run in 8+ Indian languages, and speeders, straight-liners and inconsistent answers are removed before analysis (panel methodology). The panel is in India only and app-based, so offline households are not reached.