Idea validation · Business fundamentals
How to Validate a Business Idea Before You Build It
A promising idea is a hypothesis. Validation begins when specific buyers describe the problem in their own words and take a meaningful action to solve it.
- Originally published
- September 5, 2026 by Matt Herrera
- Reviewed and updated
- September 5, 2026 by Matt Herrera
Define the decision, not the dream
Write a one-sentence hypothesis that names the customer, the costly problem, the current alternative, and the proposed outcome. Then decide what evidence would make you proceed, revise, or stop. A vague goal such as gathering positive feedback cannot protect you from confirmation bias.
The SBA recommends combining market research with competitive analysis. Use both. Market size can show that buyers exist, while direct interviews reveal whether the problem is urgent for the particular segment you can reach.
Interview around existing behavior
Ask what the person did the last time the problem occurred, what it cost, who approved the purchase, and what alternatives were considered. Avoid pitching during the interview. Compliments about a future product are weaker evidence than a recent purchase, a workaround, or a budget already assigned to the problem.
- Recruit from the exact segment in the hypothesis.
- Record patterns, objections, and the language customers use.
- Separate users, buyers, and approvers when they are different people.
- Stop or narrow the idea when urgency is consistently missing.
Run the smallest honest market test
Offer a clearly described pilot, paid diagnostic, preorder with transparent terms, or manual service that delivers the central outcome. Do not imply that a prototype has capabilities it does not have. Track qualified conversations, offers made, commitments, delivery effort, refunds, and repeat interest.
A successful test is not proof of a complete business. Feed the observed acquisition cost, price, delivery time, and retention signals into a short operating plan, then test the riskiest remaining assumption next.