Practical editorial guidance, not a promise of income or a hands-on product benchmark. Links are standard, non-affiliate links on this site. Read our editorial approach.

A promising idea is a beginning, not evidence. The next job is not to defend it. It is to find out which parts survive contact with the people you hope to help.

You do not need a finished product to start learning. You need a clear question, a relevant person, and a way to record what actually happened. Treat the sequence below as a small working experiment—not a scientific verdict on the future of a business.

Write the assumption before the pitch

Try this sentence: “I believe [specific person] struggles with [specific situation] and currently handles it by [workaround].” Mark everything you do not know. Your first conversation should help investigate those blanks.

“Small businesses need AI” is not specific enough. “A two-person design studio loses track of enquiries when project work gets busy” gives you a situation to investigate. It may turn out to be wrong. That is useful information.

Find out what happened last time. Not just what someone says might happen next time.

Ask about a real experience

Invite a relevant person to describe the most recent time the problem occurred. What triggered it? What did they try? Who else was involved? What made the current approach frustrating—or perfectly adequate?

Try questions such as: “Walk me through the last time this happened.” “What did you do next?” “What have you already tried?” and “What would you lose by changing the process?” Avoid steering every answer back to your preferred solution.

With permission, keep notes. Separate the person’s exact words, your interpretation, and your next question. A pattern you think you see is not the same as something the person said.

Look for behavior, not encouragement

Someone can like you, enjoy your idea, and still have no reason to buy it. Look for a current workaround, repeated effort, an existing purchase, a missed opportunity they can describe, or a willingness to take a specific next step.

Do not invent a pass/fail formula such as “three people liked it, therefore the market is validated.” A small set of conversations can help refine an offer, but it cannot establish market size or predict revenue.

Design one small test

Choose the question that matters most now. For a service, that might be whether a buyer will discuss a clearly scoped pilot. For an app, it might be whether someone can use a prototype to complete the intended task. For a course, it might be whether the learning outcome is specific and useful enough for a focused workshop.

Write the test before running it: who participates, what you show, what you ask them to do, what you will observe, and how you will decide whether to continue, revise, or pause. Do not quietly change the success condition after seeing the result.

Make the limits explicit

A landing page tests a message and a next action. It does not prove delivery quality. A prototype session can reveal usability problems. It does not prove that acquisition will be affordable. A paid pilot is more concrete than an encouraging comment, but one buyer does not establish a repeatable business.

Keep those distinctions in your notes. “Two people asked for a proposal” is more useful than “huge demand.” Specific evidence helps you decide what to investigate next.

Use AI to organize, not manufacture, evidence

With suitable permission and privacy safeguards, ask an assistant to organize anonymized notes into themes. Require it to distinguish direct statements from interpretations and list contradictory evidence. Review the result against the original notes.

Do not ask it to generate imaginary customer interviews and present them as research. Synthetic scenarios can help you prepare questions; they cannot tell you what actual customers need.

Decide the next move

Finish the experiment with three lines: what we observed, what remains uncertain, and what we will do next. Sometimes the right move is a narrower audience. Sometimes it is a different offer. Sometimes it is putting the idea aside.

None of those choices makes the effort wasted. The purpose of validation is to make the next decision less dependent on wishful thinking.

Turn the read into a next step.

Use the free worksheet to put this into practice. No account or email required.

Get the worksheet

Need a little help applying this? Explore coaching and implementation.