Show each participant only one concept, assigned at random

Show someone three pack designs in a row and the second gets judged against the first. By the third, they are ranking the lineup. Shoppers rarely meet your concepts that way: they see one pack on a shelf or one ad in a feed. Pure monadic testing gives each participant a single concept, so everything they say is about that concept alone.

Sixty seconds on pure monadic testing, how it works and what it changes, no sound needed.

How it works

• Add your concepts as a randomized stimulus. In the topic guide, attach a randomized stimulus to a question or a whole section and add one row per concept. Images, videos and links to YouTube, Figma, Canva, websites or PDFs all work.

• Show one per participant. Set the number of stimuli to show per participant to one. Each participant gets a single concept at random, and the questions that follow are about that concept only.

• Conveo keeps exposure close to even. Each new draw favors the concepts shown least so far, so every concept collects a similar share of interviews.

Switch between pure and sequential monadic with one number

The same setting covers both designs. Set it to one and each person sees one concept. Set it to two or more and each person sees several, one at a time, in random order, with the same questions repeated for each. That is sequential monadic, useful when your sample is too small to give every concept its own group.

Research teams asked for it in four situations

These come from teams we work with in consumer goods, food, sporting goods, toys, e-commerce and consumer apps. We kept them anonymous.

• Design variants. Three versions of a product design, each shown to its own group, so no one's view of one version is shaped by the others.

• Claims with and without. The same concept with and without a claim, in one study, with no participant seeing both.

• Positioning and messaging routes. One positioning territory or set of words per person, so each route is judged on its own terms.

• Ads and video edits. Half the sample watches one cut and half watches the other.

A common question is whether the groups stay comparable. Balancing handles the split between concepts. When the sample itself needs a set profile, add quotas.

What this means for your research

Every concept gets a first impression from people who have not seen the alternatives, the way a shopper meets one pack on a shelf. The AI research assistant then asks follow-up questions in the same video conversation, so you hear why a concept works or falls flat for each person who saw it.

Two limits to know. Balancing is approximate and does not promise equal completed interviews per concept, so watch the counts during fieldwork. And each concept needs its own sample, so pure monadic takes more interviews than showing everyone everything.

Read the stimuli guide for the full setup, including groups and parameters.