Validating a SaaS idea means reducing uncertainty about the problem, audience, value, adoption and commercial model. No single signal validates the entire business.
Map the assumptions
Write down what must be true: the problem recurs, the audience is reachable, the solution creates value and switching is acceptable. Rank assumptions by impact and existing evidence. Start with important, weakly supported assumptions rather than the easiest ones to demonstrate.
Match evidence to the question
Interviews explain a situation. Prototype tests examine understanding. Pilots observe more realistic use. A waitlist signup does not establish payment or retention. Record what each signal supports and which questions remain unanswered.
Design a test that can change your mind
Before running an experiment, define the outcomes that would alter the next decision. If every response becomes a reason to proceed, the test is not informative. Record why people decline pilots: migration effort or low urgency may explain more than favourable reactions to a demo.
A practical example
A team offers inventory tracking to independent retailers. Interviews confirm errors, but pilots reject an additional data-entry step. The problem exists; the proposed solution may need to fit the current workflow. Neither “fully validated” nor “nobody wants it” captures the evidence.
When should testing stop?
When there is enough evidence for the next decision at an appropriate risk level. Validation continues after release. One successful pilot may justify a limited version without supporting a full investment across multiple segments.
Build an evidence ladder
An enthusiastic response, a booked meeting and a completed pilot are different signals. For each, record the participant's effort and alternative explanations. Signing up may reflect curiosity. Providing representative data and involving colleagues requires more commitment, but still does not establish retention.
Imagine ten operations managers praising a purchasing-workflow idea while only two agree to show an actual purchase request. Do not convert that into a simplistic validation percentage. Investigate whether the others lack the problem, buying authority or permission to share data. Each explanation suggests a different next experiment.
Recruit beyond the enthusiasts
Include people satisfied with their current approach and people who abandoned a similar product. Ask what would make switching unnecessary. These conversations can sharpen the initial segment or prevent an expensive build. Founder communities may be accessible, but they do not automatically represent the intended buyers.
Write the decision rule before starting an experiment. Afterwards, preserve contradictory observations and withdrawals. Quietly changing the success criterion once results arrive turns validation into confirmation-seeking. A clear reason to reject an assumption is a useful product outcome.
From this guide to your project
Strategy ↗
An idea becomes an opportunity when it solves a real problem for people willing to change how they work.
Tell us about your ideaExamples are educational and do not describe client results. Content prepared with AI assistance; see our editorial policy. Our method ↗