DIY Validation Playbook
Eight steps from idea to launched product. Each step has a copy-paste prompt you can use with any AI — Claude, ChatGPT, Gemini, whatever you prefer. Do this before you write a line of code.
How to use this: Work through the steps in order. Copy each prompt, paste it into your AI chatbot, fill in the blanks, and use the output. The whole process takes 2-3 weeks — most of that time is having real conversations with people, not typing.
1
Sharpen Your Idea
15 minBefore you can validate anything, you need to describe your idea clearly. Not a pitch — a problem statement. Fill in the template below, then use the prompt to get AI feedback.
Your Idea (2-3 sentences)
What does it do? Don't describe features — describe the outcome for the user.
Who Is It For?
Be specific. "Parents" is too broad. "Parents of competitive youth athletes aged 11-14" is useful.
How Painful Is This Problem?
Nice-to-have? Painful? Hair-on-fire? Be honest.
What Do They Use Today Instead?
Spreadsheets? Word of mouth? Nothing? The current alternative tells you what you're replacing.
Once you've filled that in, paste this prompt into your AI chatbot:
I'm validating a product idea before building it. Help me pressure-test it.
My idea: [paste your 2-3 sentence description]
Target user: [who it's for]
Problem severity: [nice-to-have / painful / hair-on-fire]
Current alternative: [what they use today]
Please:
1. Tell me the strongest and weakest parts of this idea in 2-3 sentences each
2. Identify the biggest assumption I'm making that could be wrong
3. Suggest a more specific version of my target user if mine is too broad
4. Give me a one-sentence "bar napkin pitch" — if I couldn't explain this in one sentence to a stranger, it's not clear enough
2
Generate Discovery Questions
10 minYou're about to talk to real people. Not to pitch them — to learn from them. This prompt generates questions tailored to your specific idea. These are the questions you'll take into your conversations.
I need to validate a product idea by talking to real people. Generate a customer discovery script for me.
My idea: [paste your idea description]
Target user: [who it's for]
Problem severity: [nice-to-have / painful / hair-on-fire]
Current alternative: [what they use today]
Generate exactly 7 open-ended questions I should ask potential customers. Rules:
- Questions must be specific to THIS idea, not generic
- No leading questions (don't hint at the answer I want)
- Use plain conversational language — these are real conversations, not surveys
- Each question should uncover a different signal
- Include at least one question about what they've already tried
- Include at least one question about whether they'd pay
- Include at least one question about who else has this problem
Format each question with a brief note on what signal I'm looking for in the answer.
Now you have your questions. But where do you actually find people to talk to? Use this prompt to get a plan specific to your target user:
I need to find 5 people to interview about a product idea. Help me figure out exactly where to find them and how to approach them.
My target user: [who it's for — be specific]
My idea (briefly): [one sentence]
Where I live / operate: [city or "fully remote/online"]
Give me:
1. ONLINE COMMUNITIES: 5 specific places where these people already hang out (exact subreddits, Facebook groups, Slack communities, Discord servers, LinkedIn groups, forums — names, not categories)
2. IN-PERSON OPTIONS: 3 specific places or events where I could meet them face-to-face (if applicable)
3. COLD OUTREACH SCRIPT: A short, friendly DM or message I can send to someone I don't know asking for a 15-minute conversation. It should NOT pitch my idea — just ask to learn about their experience with this problem.
4. WARM OUTREACH SCRIPT: A message I can post in a community introducing myself and asking if anyone would chat for 15 minutes. Include what I'd offer in return (e.g. sharing the results, buying them a coffee).
5. WHO TO AVOID: What kind of person seems like my target user but will give me misleading data? (e.g. friends, people who are too polite, people not actually in the market)
Be specific. "Facebook groups for parents" is useless. "Facebook group: Competitive Youth Soccer Parents (127K members)" is useful.
The goal: 5 real conversations over 1-2 weeks. Not surveys, not polls — actual conversations where you listen more than you talk. Most people say yes if you ask respectfully and keep it to 15 minutes.
3
Log Your Interviews
10 min per interviewAfter each conversation, fill in this template while it's fresh. Don't wait — your memory of the conversation degrades fast. You'll need 5 of these before moving to the next step.
Who did you talk to?
Name or description. e.g. "Sarah — soccer mom, two kids aged 12 and 14"
Summary (3-5 sentences)
What did you learn? What surprised you? What confirmed what you expected?
Best quote
The single most revealing thing they said. Write it word-for-word if you can.
Pain level (1-5)
1 = shrug, 5 = they got visibly frustrated talking about this problem
Would they pay?
Yes / No / Unclear. Based on what they said and did, not what they promised.
Surprise factor (1-5)
1 = told you what you expected, 5 = completely changed how you think about the problem
Don't skip this step. The most common mistake is having the conversation and not writing it down. Unlogged interviews are wasted interviews. Do 5 before moving on.
4
Synthesize What You Heard
15 minYou've had 5 conversations. Now paste all your interview notes into this prompt and let AI find the patterns you might miss.
I just completed 5 customer discovery interviews for a product idea. Analyze my findings and give me an honest synthesis.
My original idea: [paste your idea description]
Original target user: [who you thought it was for]
INTERVIEW 1:
[paste your notes — name, summary, best quote, pain level, would pay, surprise factor]
INTERVIEW 2:
[paste notes]
INTERVIEW 3:
[paste notes]
INTERVIEW 4:
[paste notes]
INTERVIEW 5:
[paste notes]
Based on these interviews, give me:
1. REFINED PROBLEM: Rewrite my problem statement based on what people actually said (not what I assumed)
2. REAL TARGET USER: Who showed the most pain? Narrow or change my target user based on the data
3. STRONG SIGNALS: What patterns appeared across multiple interviews that suggest this is worth building?
4. RED FLAGS: What should worry me? Where did people push back or show disinterest?
5. PAIN SCORE: Average pain level across interviews (1-5)
6. WILLINGNESS TO PAY: Based on the data, would these people actually pay? Be honest.
7. RECOMMENDATION: Should I continue, pivot (and to what), or kill this idea? Give me a straight answer and explain why.
8. OPEN QUESTIONS: What 2-3 questions remain unanswered that I should investigate next?
The three possible outcomes:
Continue — strong signals, clear pain, willingness to pay. Move to scoping.
Pivot — the interviews revealed a better angle. Update your idea using what you learned and run 5 more conversations. This is the process working, not failing.
Kill — the data says no. Archive this, keep your notes, and start fresh. You just saved months of building something nobody wants.
Continue — strong signals, clear pain, willingness to pay. Move to scoping.
Pivot — the interviews revealed a better angle. Update your idea using what you learned and run 5 more conversations. This is the process working, not failing.
Kill — the data says no. Archive this, keep your notes, and start fresh. You just saved months of building something nobody wants.
5
Scope Your MVP
15 minIf the synthesis said "continue," use this prompt to turn your validated idea into a ruthlessly minimal build plan. The goal is something you can ship in 2 weeks.
I've validated a product idea through customer interviews. Now I need a scope brief for a minimum viable product I can build and ship in 2 weeks.
My idea: [paste your refined idea from the synthesis]
Target user: [paste your refined target user]
My skill level: [non-technical / hobbyist / junior developer / experienced]
Strong signals from interviews: [paste the strong signals]
Generate a scope brief with:
1. ONE-LINER: What this MVP does in one sentence
2. CORE FEATURES: 3-5 features maximum. For each, explain WHY it's essential (tie it to an interview signal). If it's not tied to something a real person said, cut it.
3. NON-GOALS: 3-5 things I am explicitly NOT building in v1. Be specific — "no mobile app" is better than "keep it simple"
4. WEEK 1 MILESTONE: What should be working by end of week 1
5. WEEK 2 MILESTONE: What ships at end of week 2 (the MVP)
6. STACK RECOMMENDATION: What tech to use for frontend, backend, database, and deployment. Factor in my skill level — don't recommend something I'll spend a week learning.
7. FIRST USER TEST: How to test this with one real user in under 5 minutes
Be ruthless about cutting scope. The #1 killer of solo builder projects is trying to build too much. If in doubt, cut it.
6
Weekly Retro
10 min / weekEvery Friday while you're building, answer these three questions. This is what separates builders who ship from builders who tinker forever.
What did I ship this week?
Features launched, conversations had, decisions made. Concrete things only.
Where did I get stuck?
Blockers, rabbit holes, things that took way longer than expected.
One thing for next week
The single most important thing. Not a to-do list — one thing.
Optionally, track these numbers each week — they compound over time:
- Cycle time — total productive hours this week
- Stuck time — hours lost to blockers or rabbit holes
- Streak — consecutive weeks you've done a retro
After 4+ weeks, paste all your retros into this prompt:
Here are my weekly retrospectives from building my product. Analyze my patterns and give me actionable advice.
[paste all your weekly retros — what shipped, what stuck, one thing for next week, and any metrics]
Based on these retros:
1. What patterns do you see in where I get stuck?
2. Am I getting faster or slower? Is my cycle time improving?
3. What should I stop doing? (Things I keep spending time on that don't move the product forward)
4. What's the one thing I should focus on for the next 2 weeks?
5. Am I building what the customers asked for, or have I drifted into building what I think is cool?
7
Price Your Product
15 minBefore you launch, you need a price. Not a guess — a number grounded in what your product costs to run, what alternatives cost, and what your interviews told you people would pay.
Help me figure out pricing for my product.
My product: [one sentence — what it does]
Target user: [who it's for]
What they use today: [current alternative and what it costs them — in money, time, or both]
What interview subjects said about paying: [paste any quotes or signals about willingness to pay from your interviews]
My costs per user per month: [hosting, API costs if using AI, tools — estimate is fine]
Give me:
1. THREE PRICING OPTIONS with a specific dollar amount for each: a "no-brainer" low price, a "fair value" mid price, and a "premium" high price. For each, explain what would justify that price to the customer.
2. WHICH MODEL FITS: Should I charge per month (subscription), per use (usage-based), or one-time? Explain why based on my product type.
3. THE MATH: At the mid price, what's my gross margin after costs? How many paying users do I need to cover $500/month in operating costs?
4. WHAT TO LAUNCH WITH: Pick one price and one model for v1. Keep it simple — I can always change it. Tell me the specific number and why.
5. ANCHORING LANGUAGE: Write me one sentence I can put on my pricing page that anchors the price against the current alternative. Format: "You currently spend [X] on [alternative]. This costs [Y]."
8
Audit Your Defensibility
10 minIf you're building an AI product, you need to know whether you're building something defensible or a wrapper that dies the moment a bigger company ships the same feature. Run this audit before you get too deep.
I'm building a product and I want to honestly assess how defensible it is. Help me run a "wrapper risk" audit.
My product: [what it does, in 2-3 sentences]
How AI is used: [describe what the AI does — is it the core product, a feature, or just a build tool?]
My data situation: [do users generate data that makes the product better? what data do I have that competitors don't?]
Integrations: [does it plug into other tools the user already uses? which ones?]
What I've built beyond the AI: [UI, workflows, templates, community, content — anything that isn't just "call an API and show the result"]
Score me on four dimensions (0-3 each, be brutally honest):
1. PROPRIETARY DATA — Do I have data that competitors can't easily get?
2. WORKFLOW INTEGRATION — Is switching away from my product painful?
3. DOMAIN EXPERTISE — Have I built specialized prompts/pipelines that would take months to replicate?
4. MULTI-MODEL RESILIENCE — Am I dependent on a single AI provider?
For each score, explain why and give me ONE specific action I could take in the next 2 weeks to improve it by one point. Then give me an overall risk assessment: Am I a wrapper, vulnerable, defensible, or strong?