Make the claim falsifiable
Write “We believe [specific group] does [observable behavior] because [constraint].” Add what would count against that belief. Avoid claims that every outcome can support. If a quiet interview means “hidden demand” and an enthusiastic one means “obvious demand,” the claim cannot help you decide.
Choose the method for the uncertainty
Use interviews to understand past behavior and constraints. Use observation to compare reported behavior with actual work. Use a prototype to examine task completion. Use a concrete commercial commitment to explore purchase intent. No single method proves the entire business model, and an attractive landing page does not establish retention.
Precommit the threshold
Before collecting results, write the sample, recruitment criteria, observable signal, and decision threshold. For a small qualitative test, the threshold is a practical decision rule rather than a claim of statistical certainty. Include a stop condition and document convenience-sampling or facilitator effects.
Review evidence and choose
Compare results with the original threshold. Separate incomplete execution from a negative result. If recruitment missed the intended buyer, repeat with the right participant rather than declaring the assumption validated. Record the rationale and the smallest next test. The aim is a better decision, not a green status badge.