How many respondents do I need for a concept test?

For directional confidence, 20 to 50 interviews from the right audience is usually enough, because every respondent produces a transcript, not just a rating. You are reading for converging objections and repeated buying language, not statistical significance.

Can I show images or a prototype?

Describe the concept in the study introduction and link to a one-pager, mockup, or video hosted wherever you like. The interview then measures the reaction to what they saw.

How is this better than a focus group?

No scheduling, no groupthink, and no moderator bias in the room. Each respondent reacts alone, in their own words, and the AI asks the same quality of follow-up at any hour in 7 languages.

What if people just say it sounds nice?

That is the answer the follow-ups exist for. The interviewer is instructed to probe vague enthusiasm: what would it replace, when would they use it, what would they pay. Politeness rarely survives three specific questions.

Templates / Concept test

Research · Adaptive interview template

Concept test template

The failure mode of concept testing is politeness. Show someone an idea and ask if they like it, and most will say yes to be nice, especially in a survey with your logo on it. The signal is not in the yes; it is in what people say when a moderator asks what they would expect it to cost, what it replaces, and what would stop them from using it.

This template pairs a simple appeal rating with an open reaction, then lets the AI interviewer do what a research moderator does: push politely on vague enthusiasm, chase specifics on objections, and ask the pricing and switching questions at the moment the respondent has just articulated their reaction.

Describe the concept in the survey introduction or link out to a one-pager, then let the interview measure the reaction.

When to use it

Before committing engineering time to a feature on the roadmap
Comparing two positioning angles for the same product
Testing a new pricing tier, bundle, or packaging change as a concept
Validating a startup idea with a target audience you can reach by link

Running it well

Test one concept per study. Comparisons belong in separate studies with matched audiences, or the themes blur together.
Recruit from the audience the concept is for, not from whoever is easiest to reach. Ten right respondents beat two hundred wrong ones.
Watch the gap between the appeal rating and the language in the open answers. Enthusiastic ratings with lukewarm words is the classic false positive.
The "I don't understand it" segment is a gift: their follow-up transcripts are a map of what your landing page will fail to explain.

The questions, and why they're shaped this way

These are the anchor questions. You can edit, reorder, or add to them; the AI interviewer handles the depth between them.

Multiple choice

Based on the description, how appealing is this to you?

I need this nowInteresting, but not urgentProbably not for meI don't understand it

The four options separate urgency from politeness. "Interesting, but not urgent" is the honest middle most forms miss, and "I don't understand it" catches messaging failures before they read as disinterest.

Open text

What would you expect this to do for you, in your own words?

The play-back question. If respondents describe a different product than the one you think you pitched, the concept has a communication problem, not a demand problem, and the transcripts show exactly where it diverges.

What the interviewer asks next

Follow-ups aren't scripted. The AI reads each answer and probes what a researcher would. Depending on what a respondent says, it might ask:

"You said you'd need this now. What are you doing today instead, and what does that cost you?"

"You mentioned it feels like a nice-to-have. What would have to be true for it to become urgent?"

"What would you expect something like this to cost per month, roughly?"

Live demo

Play the concept test interview

This is the real interviewer, not a recording. Answer as yourself and watch the follow-ups adapt to what you write.

Prefer a full window? Open the demo.

Common questions

How many respondents do I need for a concept test?: For directional confidence, 20 to 50 interviews from the right audience is usually enough, because every respondent produces a transcript, not just a rating. You are reading for converging objections and repeated buying language, not statistical significance.
Can I show images or a prototype?: Describe the concept in the study introduction and link to a one-pager, mockup, or video hosted wherever you like. The interview then measures the reaction to what they saw.
How is this better than a focus group?: No scheduling, no groupthink, and no moderator bias in the room. Each respondent reacts alone, in their own words, and the AI asks the same quality of follow-up at any hour in 7 languages.
What if people just say it sounds nice?: That is the answer the follow-ups exist for. The interviewer is instructed to probe vague enthusiasm: what would it replace, when would they use it, what would they pay. Politeness rarely survives three specific questions.

Related templates

Product

Feature validation template

Check a feature solves a real problem before you commit the roadmap.

Customers

Pricing survey template

Learn what people would actually do about a price change, and why.

Research

Market research survey template

Interview-quality depth at survey scale, in the respondent's own words.

Run the concept test with your own audience

Join the waitlist