The solo founder's idea validation checklist

You don't have a co-founder to argue you out of a bad bet or a research team to validate a good one. This checklist is the whole loop, in order, sized for one person and about three weeks of part-time work. Most items link to a longer piece if you want more detail.

Week 1: define what you're testing

• Write the idea in one paragraph. Plain language, no jargon, the version you'd say out loud.
• Name the segment precisely. "Small businesses" is too broad; "freelance designers who invoice 5+ clients a month" is specific enough. If you can't get specific, that's your first problem.
• List your assumptions. Everything that has to be true for this to work.
• Turn the top assumptions into 3 to 5 hypotheses across problem, willingness to pay, behavior, solution fit, and market. Make each one specific enough to be wrong. See how to write a startup hypothesis and startup hypothesis examples.
• Write the failure condition for each hypothesis before you collect any data. If you can't picture a plausible "no," the hypothesis is still too vague.
• Rank hypotheses by risk. Test the one the whole idea depends on first.
• Draft a screener question that disqualifies people who don't have the problem.
• Draft 8 to 12 questions about behavior and past spend. Pull from customer discovery interview questions.
• Check every question for leading language. Remove any mention of your solution until the problem is established. See the Mom Test rewrites.

Week 2: collect evidence

• Find where your segment gathers: subreddits, Slack and Discord groups, LinkedIn, niche newsletters.
• Recruit 10 to 20 qualified people for round one. Fit matters more than volume. See how many customer interviews you need.
• Run the screener first and politely end conversations with people who don't qualify.
• Ask about the last specific time the problem came up, every time.
• Follow up three times on every interesting answer: why was that a problem, what did you do, what happened next.
• Record everything (with permission). You'll need exact quotes later.
• Ask for referrals at the end of every conversation.
• If you're running a survey instead, keep it under 12 questions, put the screener first and the open-text questions early. See idea validation survey questions.
• Test willingness to pay against real past spend instead of a hypothetical number. See willingness to pay survey questions.

Week 3: read the verdict

• Score each hypothesis as confirmed, weakened, or inconclusive, based on what people actually reported.
• Check your scoring against the bar you set in week 1. Don't move the goalposts now.
• Look for the stop signals: mild problem language, no existing workaround spend, no recent instance, purely hypothetical willingness to pay.
• Roll the hypotheses into one read: build it, refine it, keep testing, or weak demand.
• If it's a kill, write down which assumption broke. That's what you'd change on another attempt. See when to kill a startup idea.
• If it's "refine," change one variable (segment, price, or framing) and run another focused round.
• If it's "build," write a one-line summary of the evidence so you can point to it when you doubt yourself in month three.

A few things to keep in mind

• A clean "no" in week three is a win. It hands you back five months.
• You're the least objective person about this idea, so decide your criteria before the data arrives.
• Enthusiasm from friends doesn't count as evidence. Behavior from strangers does.
• One good screened respondent beats ten unscreened ones.

Common questions

Can I really validate an idea in three weeks?

For one hypothesis-testing round, yes. Some ideas need two or three rounds, especially if the first is ambiguous. Getting a build-or-kill signal on the core problem takes about three weeks, though, and three months is far more than you need.

What if I've already started building?

Run the checklist anyway, starting from week 1. Confirming a weak idea now costs a few weeks. Finding out after launch costs everything in between.

Do I need a budget?

Not really. A small budget for a recruiting panel speeds up week 2, but relevant communities and direct outreach cost nothing.

What's the most-skipped step?

Writing the failure condition before collecting data. It's the one step that stops a weak result from being rationalized into a green light.

Run the whole checklist in one place

Hypothis walks a solo founder through this loop: hypotheses, a screened research round, adaptive follow-ups, and a scored verdict that's allowed to say no. It's free during early access. See how it works or create your account.