# The AI-Powered 72-Hour Kill Switch

> Three-day prompt framework to validate or kill startup ideas fast

A 72-hour idea validation framework designed by Tom Bilyeu that uses AI prompts to test business concepts before building. It breaks the validation process into three 24-hour phases—pain research, market analysis, and live demand testing—paired with explicit, measurable kill criteria to stop unviable projects early.

## The framework

Phase 1 — Hour 1-24: AI Pain Research. Feed ChatGPT: "Search Reddit, forums, and review sites for people complaining about \[your problem\]. Find 50+ complaint threads from the last 6 months. Look for phrases like 'there has to be a better way' or 'I hate how \[current solution\] works.' Give me direct quotes and links." Then ask: "Analyze these complaints. What specific words do people use? How urgent does their frustration sound? Rate the pain level 1-10." Kill criterion: no real complaints with high urgency = no real problem / dead idea.

Phase 2 — Hour 25-48: AI Market Analysis. Prompt sequence: "Research companies solving \[your problem\]. Find their pricing, customer reviews, and revenue estimates. I need: 3+ companies charging $50+ monthly; customer testimonials mentioning ROI; evidence people are actually paying these prices; market size estimates." Follow-up: "Analyze their customer reviews. What do people love? What do they hate? Where are the gaps I could fill?" Kill criterion: no money flow = dead idea.

Phase 3 — Hour 49-72: AI Content Creation & Testing. Landing page prompt: "Create a landing page for \[your solution\]. Include: headline that speaks to the specific pain points you found; 3 bullet points of benefits; email signup form copy; social proof placeholder text. Write it for someone who just complained about this problem on Reddit." Social posts prompt: "Write 5 different social media posts announcing my 'coming soon' product. Make each one sound like a real person sharing their frustration, not a sales pitch. Include the signup link naturally." Post across LinkedIn, Facebook groups, and Reddit threads where people complained; track signups per post. Kill criterion: under 2% signup rate = dead idea.

BONUS — AI Survey Analysis: email signups with AI-generated copy: "Draft a survey email asking my signups what they'd pay monthly to solve \[problem\]. Make it conversational, not corporate. Include 3 follow-up questions about their current workarounds." Feed responses back to AI to find the average. Kill criterion: average willingness to pay under $50 = a hobby, not a business.

Closing principle: "Speed kills bad ideas. AI kills them faster."

## What it is for

- Validate customer pain and demand for a new product concept in three days
- Identify competitor weaknesses and pricing gaps using AI-assisted market research
- Draft landing page copy and social posts tailored to existing online complaints
- Assess pricing power and customer willingness to pay using conversational survey copy

## Key points

- Phase 1 (Hours 1–24): Pain research prompts to mine forums and reviews for urgent user frustrations
- Phase 2 (Hours 25–48): Competitor and market research prompts to verify pricing and existing cash flow
- Phase 3 (Hours 49–72): Content generation prompts for landing pages and social posts to measure signups
- Survey analysis bonus: AI-drafted follow-up email to quantify average willingness to pay
- Predefined kill criteria at each phase to eliminate bad ideas without emotional attachment

## Pricing

Free

## Details

- Type: Framework
- Category: [Business & Money](https://xiry.app/topics/business-money.md)
- Topic: [Idea Validation](https://xiry.app/topics/business-money/idea-validation.md)
- Open source: no
- Platforms: other
- For: Founders, solopreneurs, and product builders testing new business ideas
- Alternative to: The Lean Startup framework, Strategyzer Customer Discovery
- Tags: frameworks, prompts, startup, idea validation, chatgpt, entrepreneurship

## Why it was saved

- [@TomBilyeu, 30 June 2025](https://x.com/TomBilyeu/status/1939723238214381890): Lets an entrepreneur test a business idea in 72 hours instead of 12 months, with concrete prompts and measurable kill criteria.
- [@TomBilyeu, 30 June 2025](https://x.com/TomBilyeu/status/1939723236180173191): Kill bad ideas in 3 days instead of months of building, by checking pain, market, demand, and willingness to pay with AI.

## Related

- [60-Minute Business Idea Validation Script](https://xiry.app/resources/60-minute-business-idea-validation-script): A timed, five-step framework to test and validate business ideas rapidly. (Tip, Free)
- [AI Market Analysis Prompt](https://xiry.app/resources/ai-market-analysis-prompt): Two-step prompt to evaluate competitors, pricing, and market demand for an idea. (Prompt, Free)
- [AI Pain Research Prompt](https://xiry.app/resources/ai-pain-research-prompt): Prompt sequence to mine customer complaints and validate problem urgency. (Prompt, Free)
- [B2C App Idea Filter](https://xiry.app/resources/b2c-app-idea-filter): A nine-question checklist for validating consumer app ideas before building (Framework, Free)
- [Idea Validation Prompts for Landing Pages and Social Posts](https://xiry.app/resources/idea-validation-prompts-for-landing-pages-and-social-posts): AI prompts for generating validation landing pages, teaser posts, and survey emails (Prompt, Free)
- [Pre-Launch Assumption Testing Framework](https://xiry.app/resources/pre-launch-assumption-testing-framework): A method to validate core business assumptions before writing code or spending capital. (Tip, Free)

Updated 11 October 2026. Page: https://xiry.app/resources/the-ai-powered-72-hour-kill-switch

Make your own library: Xiry takes your saved reels, tweets, Reddit bookmarks and Watch Later videos, makes them searchable and pulls the useful stuff out, like this library. It's invite-only for now; join the waitlist at https://xiry.app/#waitlist.