How to Validate a Startup Idea Before You Start Building

What 'validating an idea' actually means

Validation is not asking friends if your idea is cool. It is generating evidence that a specific group of people has a specific painful problem and is willing to pay, sign up, or take a costly action to solve it.

The three signals that matter: they can describe the problem in their own words, they have already tried other solutions, and they will commit money, time, or their email in exchange for a fix.

Step 1 — Write a one-sentence hypothesis

Fill in the blanks: '[Specific person] struggles with [specific problem] when [specific situation], and today they solve it by [current workaround].' If any blank is fuzzy, you do not have a hypothesis yet — you have a wish.

Step 2 — Run 8-12 problem interviews

Interview people who match your target. Ask about the last time the problem happened, what they did, what it cost them, and what they wish existed. Do not pitch. If you cannot find eight people to talk to, that is your first signal.

Step 3 — Test demand with a smoke page

Put up a one-page site describing the outcome, not the product. Drive 200-500 targeted visitors via a small ad spend, a niche subreddit, or a founder community. Measure the sign-up rate. Below 3% on cold traffic is a weak signal; above 8% is worth exploring further.

Step 4 — Score the idea against a rubric

Rate the opportunity across five dimensions so you compare it against your other ideas objectively:

• Pain severity (1-5): how urgent is the problem?
• Market pull: are people actively searching for a solution?
• Willingness to pay: will they commit money, not just interest?
• Distribution: do you have a repeatable way to reach them?
• Founder fit: do you have unfair insight or access here?

Step 5 — Decide: build, pivot, or drop

Anything scoring below 15/25 is a drop. 15-19 is a pivot signal — usually a narrower audience or a different job. 20+ is a green light to plan an MVP.

Run this loop on every idea. The founders who ship the right product first are the ones who kill wrong ideas fastest.

How Forge AI accelerates this

Forge AI runs the interview script, competitor sweep, demand-signal analysis, and scoring in one pass so you get a validated brief instead of a blank page. Start with your idea and get a research report in under two minutes.