How to Validate a Startup Idea: A Practical Evidence Checklist
Validate a startup idea by collecting evidence that a specific customer has a worthwhile problem, can be reached, and will take a meaningful step toward a solution. The useful output is a decision about what to test next, supported by observations. A folder of encouraging comments is not enough.
Name one customer segment and the decision your next test should resolve.
Separate problem evidence, solution interest and willingness to pay.
Record failed tests and alternative explanations alongside positive signals.
Make the next investment proportional to the strength of the evidence.
1. Turn the idea into a testable claim
Write a one-sentence hypothesis before making a landing page: a particular customer struggles with a particular task under particular conditions. Add the workaround they currently use and why you believe it is inadequate. This makes disagreements visible. Two founders may agree that scheduling is frustrating while imagining entirely different buyers, budgets and workflows.
For example, consider a hypothetical tool for independent tutoring businesses. The starting claim could be that owners handling several tutors spend too much time rearranging cancelled lessons. That is more useful than saying every education business needs an AI assistant. List what would change your mind: infrequent cancellations, a satisfactory existing tool, or no authority to change the process.
2. Check the market without treating research as proof
The SBA describes market research as a way to investigate demand, alternatives, location and pricing. It also distinguishes existing research from direct research with customers. Use that distinction to assemble context before making claims about your own offer. Read the SBA market-research guidance.
Create a short alternatives sheet with actual products, manual workarounds and the option of doing nothing. Record the source and date for every price. Look for constraints that could prevent adoption, such as a school-approved software list or a parent communication policy. A large category does not demonstrate that your specific segment is reachable or dissatisfied.
3. Match the experiment to the uncertainty
Strategyzer separates testing customer problems, testing an offered solution, and testing willingness to pay. These are different questions, so evidence for one should not silently stand in for another. See Strategyzer’s testing roadmap.
Use a recent-event interview to investigate how cancellations are handled. Use a simple workflow demonstration to see whether an owner understands the proposed process. Use a clearly described pilot invitation to examine commitment. Be explicit about what exists and what is experimental. Do not collect payments for an unavailable service without clear delivery and refund arrangements.
Problem test: ask someone to reconstruct a recent incident and show the workaround.
Solution test: observe a realistic task instead of presenting a feature tour.
Commitment test: ask for a scheduled pilot, a responsible owner or another proportionate next step.
Channel test: record how suitable participants were reached and the effort required.
4. Use a decision sheet: a hypothetical tutoring pilot
Suppose the team interviews owners and learns that cancellation frequency varies sharply. Some describe weekly rescheduling, while others already use a satisfactory booking system. The appropriate response is to narrow the segment to businesses with the relevant workflow. It is not to combine all reactions into an average satisfaction score.
For the next experiment, invite suitable owners to try a manually supported rescheduling process for one agreed period. Define an observable task, record the original workflow, and agree what information may be used. Compare the result with the previous process. A pilot that works only because a founder spends hours on each change may establish a useful need while leaving delivery economics unresolved.
Decision |
Evidence to capture |
What remains unknown |
|---|---|---|
Is the problem worth pursuing? |
Recent incidents, current workaround and consequences |
Whether owners will change tools |
Does the proposed process help? |
Completed task and observed friction |
Whether it works without intensive founder support |
Is a commercial test justified? |
Explicit next-step commitment from the buyer |
Repeat use, sustainable price and delivery cost |
5. Decide what the evidence allows you to do
Finish each test with one of four decisions: continue, narrow, change or stop. Include the observation that drove the decision and the strongest competing explanation. A missed pilot may indicate low urgency, but it could also reflect a busy season or an unclear invitation. Investigate the explanation before redesigning the whole business.
Keep one dated evidence log instead of rewriting the story after each conversation. When the core uncertainty becomes whether users can complete a valuable task, move to MVP planning. For the interview mechanics, use the customer-discovery guide. NextUnicorn describes Idea2Unicorn as supporting idea validation and MVP planning; its program page is the place to assess that offer. Explore Idea2Unicorn.
Frequently asked questions
There is no universal number. Start with a defined decision and a narrow segment, then look for consistent evidence and meaningful contradictions. Interviews help explain behavior; they do not by themselves establish payment, retention or delivery economics.
Yes. Interviews, workflow demonstrations and transparently described manual pilots can test several assumptions. Use the smallest method that answers the question, while being honest about what is already available.
Positive feedback is a starting signal. Ask what behavior supports it: a recent problem, an existing workaround, a completed trial or a concrete next step. Build when the next learning question actually requires a product.
Explore Idea2Unicorn’s published support for idea validation and MVP planning.
- Plan your business: market research and competitive analysis - U.S. Small Business Administration
- Roadmap to test your value proposition - Strategyzer
- Idea2Unicorn program - Next Unicorn Fund