THE SHORT ANSWER
Start from a research question, recruit relevant people and ask neutral questions about recent real behavior: when the situation last occurred, what they did, what was difficult and what happened next. Treat statements as evidence to analyse, not automatic proof of demand.
Prefer events over predictions
| Avoid relying on | Ask instead |
|---|---|
| Would you use this? | How do you solve this problem today? |
| Would you pay ₹500? | What did the last solution cost in money, time or effort? |
| Do you like this idea? | When did this problem last happen? What did you do next? |
| Is this feature useful? | Which step was difficult, and what happened because of it? |
Build a flexible discussion guide
Define what the research needs to learn, then prepare topic prompts and possible follow-ups. Test the wording with a colleague. During the session, listen for real examples and allow relevant evidence to change the order.
Obtain informed consent and protect personal data. The fact that a question could be useful does not make it appropriate to ask or retain.
Evidence & context: GOV.UK Service Manual
Do not teach the participant your answer
Wording and order can influence responses. Keep prompts neutral, avoid praising only confirming answers and separate observation from your interpretation. One vivid interview is not a market estimate.
Use the deeper Startup Idea Validation process to connect interviews with hypotheses and other evidence, and User Journey Product Design to translate findings into a journey.
Evidence & context: Pew Research Center
Rewrite five research questions
- Remove the proposed solution from the question.
- Anchor the prompt in a recent event.
- Ask what happened before and after.
- Add one neutral probe for difficulty or consequence.
- Write what this answer can and cannot establish.
Sources & further reading
- Using In-Depth Interviews
GOV.UK Service Manual. Practical public-service guidance on research questions, neutral prompts, follow-up questions, consent and listening. It is professional guidance rather than experimental evidence and must be adapted to the research context.
- Writing Survey Questions
Pew Research Center. Methodological guidance showing how open or closed formats, wording and question order can affect responses. Survey design differs from ordinary conversation, but the examples demonstrate why question form is consequential.
Examples and exercises are illustrative unless attributed to a source. No independent expert review is claimed.
A correction, a counterexample or an experience worth sharing?
Join the conversation ↗