Skip the Spreadsheet: Test Product-Market Fit With Real Money, Not Opinions
Most people who say they 'did market research' really just browsed the internet until they found a few articles that confirmed what they already believed. Nothing you read at your desk tells you whether someone will actually pull out their wallet. The surface problem is 'I don't know if there's demand'; the real problem is that people conflate what someone tells them with what someone pays for, and never actually test the difference. This prompt doesn't build a desk analysis, it builds a field validation test anchored in real buying signal: a presale or waitlist mechanic, a structured five-question interview protocol, and concrete thresholds that turn a fuzzy feeling into a go or no-go decision. The goal isn't to convince you the idea works. It's to measure whether the market is convincing you.
You are an early-stage startup validation specialist. Your expertise: testing an idea in the field with real buying signal, not at a desk. Your job is NOT to validate or debunk the idea; it's to design a test that collects a decision-ready signal from real people who vote with money. I'll give you: - [IDEA: what I want to sell, what problem it solves] - [MY ACCESS CHANNEL: who I can actually reach, which group/platform/network] - [BUDGET: what I can spend on this test, may be close to zero] - [TIME: how many weeks I want to finish this test in] - [WHAT I'VE DONE SO FAR: any previous steps I've already taken on this idea, if applicable] RETURN THE OUTPUT IN EXACTLY THIS ORDER, WITH THESE EXACT HEADERS: ═══ PRESALE / WAITLIST MECHANIC ═══ A simple mechanic where people leave a real commitment (name, contact, position in line) before the actual product exists. You can get a serious signal without collecting card details or money. Specify how this mechanic gets shared through the access channel given in the input. ═══ 5-QUESTION CUSTOMER INTERVIEW PROTOCOL ═══ Five questions, in order. Probe current behavior and frustration first, introduce the idea last, and put the price question at the very end. Briefly explain why each question sits where it does. ═══ SIGNAL TYPES TO COLLECT ═══ Which answers count as "real demand" and which count as a "polite no"; concrete tells for telling them apart. Explain how to separate someone saying "yeah, that's interesting" to be nice from someone actually getting in line. ═══ GO / NO-GO THRESHOLDS ═══ Out of how many people, how many need to show which behavior to move forward, and how many mean you change the idea or the angle. Numeric and specific; a realistic threshold scaled to the size of the access channel in the input. ═══ NEXT STEP IF THE THRESHOLD ISN'T MET ═══ If the test fails, which variable to change (price, audience, pitch angle) and retest before abandoning the idea entirely. State why rerunning the identical test on the identical audience is a waste of time. ═══ COMMON MISTAKES THAT RUIN THIS TEST ═══ The 2-3 most common ways people run this kind of test wrong (asking friends and calling it market research, asking the price question too early, reading their own excitement as the answer) and how to catch each one. RULES: Don't use made-up conversion rates or a made-up "typically X% of people sign up" statistic. Size the test to the real access channel given in the input; don't assume a large imaginary audience. When collecting personal data (name, phone number), suggest language that respects consent and transparency. Don't break the order of the interview questions; the price question always stays last. Write in clear, plain, actionable English.
How to use it
- Write down your idea and your most realistic access channel, exactly who you can reach and where (which group, platform, or network you actually have standing in). The test's honesty depends on real reach; you can't run this against an imaginary audience.
- Follow the 5-question interview protocol in the output word for word, especially the order of questions before the 'how much would you pay' question. Break the sequence and the answer collapses into a polite 'yeah, I'd buy that', and the real signal disappears with it.
- When you hit the go or no-go threshold, don't decide on a feeling, look at the number the output gives you. If the threshold isn't met, change your angle, not your idea, the product usually isn't the problem, who you're pitching it to and how is.
Example / tip
Input: Idea = a weekly fresh dog food subscription box for pet owners, planning to launch locally first. Access = member of 3 dog owner Facebook groups, and know a local vet clinic. AI output (shortened): Presale mechanic, before the real product exists, an 'early signup for the first 20 boxes, no card required, just intent' form. Interview protocol: 1. What are you feeding your dog right now, 2. What's your biggest complaint about it, 3. What have you tried to fix that, 4. (describe the box) what do you think of this, 5. What would you pay monthly, don't ask for card details, just get them to say a number out loud. Go threshold: if at least 8 out of 20 people leave a real name and phone number on the waitlist, move forward; under 3, change the angle.
Usage example
Copy the prompt as is, fill in the bracketed fields for your own business, and paste it into ChatGPT or Claude. The more concrete your input, the more useful the output. Take the first draft, ask for a one-sentence fix on whatever you don't like, then give it a final pass in your own voice.
When not to use it
This prompt sets up early validation for a small-scale audience you have direct access to; it doesn't replace a professional market research firm for decisions that require large-scale investment. If you're taking presale money or deposits tied to a product commitment, check consumer protection law in your jurisdiction separately.
Output quality checklist
- Does the interview question order probe real behavior before the price question, or does it just ask 'would you buy this' outright?
- Is the go/no-go threshold tied to a concrete number, or does it stay vague, like 'if I get good feedback'?
- Does the presale mechanic ask for a real commitment (name, contact, position in line), or does it just collect likes?
- Is it customized to the access channel given in the input, or does it hand back a generic template?
