THE SHORT ANSWER
Validate a startup idea by naming the most consequential assumption, choosing the smallest ethical test that could challenge it, observing relevant customer behaviour and deciding what the result changes. Interviews support discovery; stronger commitments such as time, data, usage or payment can test demand more directly.
Separate interest from evidence
| Method | What it can reveal | Important limitation |
|---|---|---|
| Interview | Context, language and past behaviour | Stated intent is not future action |
| Observation | Actual workflow and friction | One setting may not generalize |
| Landing page or waitlist | Response to a defined promise | A signup is not retained usage |
| Prototype or pilot | Whether the solution helps | Facilitation may inflate success |
| Manual delivery | Demand and workflow before automation | Delivery economics may later change |
| Pre-order where appropriate | A stronger commitment to buy | Refunds, disclosure and fulfilment matter |
Evidence & context: Strategyzer
Ask about events, not compliments
‘Do you like my idea?’ invites politeness and imagination. Ask instead about the last relevant event, the current alternative, who decided, what was spent and why the person acted or did not act.
Design one learning loop
- Assumption: what must be true?
- Risk: why would being wrong matter?
- Test: what is the smallest credible experiment?
- Signal: what observable result will be recorded?
- Threshold: what result changes confidence?
- Decision: continue, revise, stop or test again?
Set the interpretation before seeing the result where possible. Moving the threshold afterwards makes it easier to preserve a preferred story.
Evidence & context: Strategyzer
Validation reduces uncertainty; it does not remove it
A waitlist can overstate demand, a pilot can depend on founder effort and a pre-order can still be cancelled. Record who participated, the conditions, the sample, the observed behaviour and what remains unknown.
Once the problem signal is credible, choose the smallest useful MVP that tests the next uncertainty.
Sources & further reading
- Business testing: is your hypothesis really validated?
Strategyzer. Practitioner guidance distinguishing directional discovery evidence from stronger evidence of real-world behaviour. Its evidence scale is a method, not a guarantee.
- Designing strong experiments
Strategyzer. Practitioner guidance on explicit hypotheses, relevant participants and well-designed artefacts. Experiment quality and interpretation still depend on context.
- Market research and competitive analysis
U.S. Small Business Administration. Official planning guidance on demand, market size, location, saturation and pricing. It is a research framework, not proof that a particular opportunity will succeed.
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 ↗