Concept Testing Interview Guide: Test Understanding Without Selling the Idea

A concept testing interview guide helps a team learn how people understand an early idea before internal language and enthusiasm harden into assumptions. The interview should reveal comprehension, relevance, concerns, missing information and tradeoffs without coaching participants toward approval.

Concept testing is evidence for a decision, not a referendum. A positive reaction does not prove demand, usability, feasibility or commercial success. State what the study can and cannot support.

Define the decision the test will inform

Write the decision first: whether to refine a value proposition, choose what to prototype, compare two framings or identify critical unknowns. Then define the research questions that provide evidence for that decision.

Avoid a vague objective such as “see what users think.” Use questions such as: What does the participant believe the concept does? In which real situation might it matter? What creates doubt? Which existing behavior would it need to replace?

The decision owner and research owner should agree in advance what evidence would cause refinement, further research or a stop. The interview itself does not authorize a launch.

Recruit relevant participants and contexts

Recruit people whose recent experience matches the decision. Screen for the situation, behavior or constraint the concept addresses, not merely general interest in the category. Record important exclusions and sample limitations.

Power relationships, incentives and prior exposure can shape answers. Separate employees, customers, managers or channel partners when those relationships would suppress disagreement. Explain compensation consistently and never make it dependent on a favorable response.

Use user interview questions to build neutral screening and experience prompts. Do not overstate a convenience sample as representative of an entire market.

Prepare a controlled concept stimulus

Use the least polished stimulus that communicates the idea accurately enough for the research question. A paragraph, storyboard, mock screen or simple prototype may be appropriate. High visual polish can create confidence unrelated to the concept; low fidelity can create confusion unrelated to the intended value.

Keep mandatory information consistent across participants. Version every stimulus and record presentation order. Mark placeholders and hypothetical behavior. Do not present unbuilt functionality as currently available.

Pilot the stimulus with someone outside the core team to find accidental explanations, jargon and visual cues that reveal the preferred answer.

Copy this concept testing interview guide

CONCEPT TEST INTERVIEW

Study / concept version:
Decision and research questions:
Participant reference:
Moderator / note-taker:
Consent and recording status:

1. OPEN — 5 MIN
Purpose, rights, confidentiality limits and session format

2. CURRENT EXPERIENCE — 15 MIN
Last relevant situation, trigger, process, alternatives and outcome

3. UNAIDED CONCEPT RESPONSE — 15 MIN
What is this? Who is it for? What might happen next?

4. RELEVANCE AND EVIDENCE — 15 MIN
Where could it fit? What would it replace? Why or why not?

5. CONCERNS AND TRADEOFFS — 10 MIN
Questions, trust, effort, risk, missing information and exceptions

6. COMPARISON — 10 MIN
Alternative concept or current approach; reasons, not votes

7. CLOSE — 5 MIN
Recap, correction, missing perspective and next steps

Use optional probes selectively. Completing every question is less important than understanding a relevant contradiction in depth.

Begin with current behavior

Before showing the concept, ask about the participant’s most recent relevant experience. Explore the trigger, steps, tools, people, workarounds and outcome. Concrete recall provides a baseline against which the concept can be discussed.

Ask what they currently do, not what they claim they would ideally do. Hypothetical enthusiasm is easy when no real tradeoff is present. If the participant has never faced the intended situation, that itself is useful screening evidence.

Do not name the concept’s promised benefit in the warm-up. That language can prime the participant to repeat it later.

Capture unaided understanding first

Present the controlled stimulus without a sales explanation. Allow silence. Then ask:

• What do you think this is?
• What do you think it would help someone do?
• Who seems to be the intended user?
• What would you expect to happen first?
• Which words or parts are unclear?
• What questions would you need answered?

Record the participant’s own terms before clarifying. If they misunderstand, explore what in the stimulus produced that interpretation. Correct factual confusion only after the unaided response is complete and mark the point where additional explanation began.

Test relevance through evidence, not praise

Ask the participant to connect the idea to a real situation: “When, if ever, would this have been useful in the example you described?” Follow with what they would do differently, what effort it would require and what current option it might replace.

Probe weak signals. “That sounds useful” may be politeness. Ask what specifically creates value, what would stop action and what evidence they would need before trying it. Do not ask participants to predict purchase, adoption or long-term behavior with false precision.

The customer churn interview guide demonstrates how concrete episodes and alternatives provide stronger evidence than generalized satisfaction questions.

Explore concerns, trust and tradeoffs

Invite criticism explicitly: “What would make this a poor fit?” Explore effort, privacy, learning, reliability, workflow disruption, social consequences and missing control as relevant to the concept. Do not rebut concerns during the research portion.

Ask when the concept should not be used and who might experience it differently. Negative cases reveal boundaries that average reactions conceal. If a concern requires specialist interpretation, record it and route it to the qualified owner rather than improvising an answer.

Distinguish a participant’s question from a verified product requirement. Research evidence informs prioritization; it does not automatically set scope.

Compare concepts without forcing a vote

If comparing concepts, rotate or randomize presentation order where practical and document the method. Capture an unaided response to each before asking for comparison. Then explore which situation makes one more appropriate and why.

A preference count from a small qualitative sample is not a market forecast. Preserve the reasons, contexts and contradictions behind choices. One concept may communicate clearly while another addresses the more important need after explanation.

Use a research debrief template after sessions to compare evidence while retaining counterexamples and source references.

Manage consent, recording and sensitive moments

Explain purpose, participation rights, recording, access, retention, confidentiality limits and withdrawal or correction routes according to the approved research process. Collect only information needed for the study. Stop probing when a participant declines or when the topic moves outside the authorized scope.

For an authorized concept interview with visible, agreed capture, Kuno can help create a draft transcript, notes and follow-up actions for researcher verification. It does not infer consent or decide whether the concept succeeds. Explore Kuno

If distress, safeguarding, privacy or material product-risk concerns arise, follow the predetermined escalation route. The moderator should have an available owner and contact method, not search for one during the session.

Synthesize evidence with owners and escalation

After each session, separate observed behavior, participant statement, moderator interpretation and team implication. Link important evidence to the session reference and stimulus version. The qualitative research transcription quality checklist helps verify key quotations and speaker context before analysis.

Assign owners to open questions. Escalate when a concept appears to create material harm, the stimulus misrepresents current capabilities, required consent or data controls failed, or stakeholders pressure the team to suppress contradictory evidence.

Decision owners should receive findings, limitations and unresolved questions together. A recommendation without its evidence boundary encourages false confidence.

Complete human review and final QA

Before reporting, confirm that the sample matches the stated criteria, all sessions used the correct stimulus versions and material findings trace to reviewed evidence. Check that minority views and negative cases remain visible and that later explanations are not confused with unaided comprehension.

Verify quotations, remove unnecessary identifying detail, document researcher influence and state what the study cannot conclude. The research owner reviews the analysis; relevant subject owners review specialist implications; the decision owner decides the next step.

Review the final report against the original decision and research questions. Remove observations that are interesting but unsupported or outside scope, unless they are clearly labeled as follow-up questions. Check that the report does not count repeated comments as independent proof or collapse different participant contexts into one claim. If the team changes the concept after testing, record which findings still apply and which need new evidence. A decision to proceed should state assumptions, safeguards and the next validation point rather than treating one study as permanent confirmation.

Use conversation capture to preserve evidence, not to automate product judgment. Kuno supports authorized research drafts while accountable people verify the record and own the decision. See Kuno