# Would it work? (wouldit.work)

> An open-source idea check for vibe coders and AI-agent builders. Before you build, see who already built it and what real users complain about. Every claim links to its source.

Would it work? has two parts: (1) an idea checker that searches GitHub and Hacker News for prior art and real complaints and writes a report where every claim links to its source; (2) a library of sourced case studies of real indie, AI and vibe-coded products.

Key facts:
- The idea check is free for 3 checks per day; individual reports are private (noindex) and never listed publicly.
- Evidence sources: GitHub repositories and Hacker News stories and comments. When evidence is thin, the report says "Not enough data".
- Every number in a case study links inline to the page it came from; self-reported figures are labelled as such.
- Full text of every case study: https://wouldit.work/llms-full.txt

## Product
- [Check your idea](https://wouldit.work/check): free, 3 checks per day; reports are private (noindex).
- [Pricing](https://wouldit.work/pricing): Free; Pro US$9/month auto-renewing (cancel anytime); Single deep check US$3 one-time.
- [About](https://wouldit.work/about)
- [Contact](https://wouldit.work/contact): support@wouldit.work

## Case studies
- [TypingMind: a weekend ChatGPT front-end that grew past $1M a year](https://wouldit.work/case-studies/typingmind): How Tony Dinh turned a better UI for the ChatGPT API into a $1M+/year product, starting with one-time licenses and adding team subscriptions later.
- [fly.pieter.com: a vibe-coded game that hit '$1M ARR' in 17 days, then $0](https://wouldit.work/case-studies/fly-pieter): Pieter Levels built a browser flight simulator with AI and sold in-game ads and planes. The headline was $1M ARR in 17 days. The follow-up shows why one-off revenue is not ARR.
- [Cal AI: a photo calorie counter that MyFitnessPal bought within two years](https://wouldit.work/case-studies/cal-ai): Two high-school founders shipped a 'snap your food' calorie app in 2024. It passed $30M in annual revenue and was acquired by MyFitnessPal.
- [ShipFast: a Next.js boilerplate built in a week that made $6,000 in 48 hours](https://wouldit.work/case-studies/shipfast): Marc Lou packaged code he kept rewriting for his own apps into a boilerplate. It made $6K in two days and $220K in about four and a half months.
- [TrustMRR: built in under 24 hours as a reply to one tweet, $20,378 in 3 days](https://wouldit.work/case-studies/trustmrr): Marc Lou built a verified-revenue leaderboard after Pieter Levels complained about fake MRR screenshots. Ad slots sold out in three days.
- [Tally: a free form builder that grew to $5M ARR, fully bootstrapped](https://wouldit.work/case-studies/tally): Tally launched in 2020 with a generous free tier and a 'Made with Tally' badge. Five years later it passed $5M ARR with 11 people.
- [Plausible Analytics: $1M ARR with no ads, by picking a fight with Google](https://wouldit.work/case-studies/plausible): A two-person open-source analytics tool took 324 days to reach $400 MRR, then grew to $1M ARR through blog posts that hit the Hacker News front page.
- [HeadshotPro: AI headshots, $10M+ in revenue, with affiliates as a growth engine](https://wouldit.work/case-studies/headshotpro): Danny Postma launched HeadshotPro in 2023 and grew it into an AI headshot studio with $10M+ lifetime revenue. Affiliates bring in $50K+ a month.
- [Photo AI: Pieter Levels' AI photo studio built on plain PHP and jQuery](https://wouldit.work/case-studies/photo-ai): How Photo AI went from launch in February 2023 to $61,808 per month by July, with a tech stack most developers would call outdated.
- [Interior AI: built in 5 days, later reported at ~$45K MRR with >99% margins](https://wouldit.work/case-studies/interior-ai): Pieter Levels built Interior AI in five days in 2022. In 2024 he reported about $45,000 MRR and around $250 in monthly costs. Here is what that teaches, and what the skeptics said.
- [Formula Bot: a weekend no-code app, one Reddit post, and a $4–5K API bill](https://wouldit.work/case-studies/formula-bot): David Bressler built Excel Formula Bot in Bubble over a weekend. A Reddit post brought 100,000 visitors overnight and a surprise OpenAI bill before pricing was in place.
- [Bannerbear: two years, three pivots, then a steady climb to $50K MRR](https://wouldit.work/case-studies/bannerbear): Jon Yongfook burned a year of savings on 12 startups in 12 months, narrowed to image generation, pivoted to an API, and grew with a strict code-one-week, market-the-next rhythm.
- [GummySearch: 140,000 users, then shut down by a platform's API policy](https://wouldit.work/case-studies/gummysearch): A solo-built Reddit audience-research tool earned a large following, then closed in 2025 after failing to get a commercial Reddit API license. A lesson in platform risk.

## Case studies by category
- [AI tools](https://wouldit.work/case-studies/category/ai-tools): 5 case studies
- [SaaS](https://wouldit.work/case-studies/category/saas): 3 case studies
- [Developer tools](https://wouldit.work/case-studies/category/developer-tools): 2 case studies
- [Vibe-coded](https://wouldit.work/case-studies/category/vibe-coded): 2 case studies
- [Consumer apps](https://wouldit.work/case-studies/category/consumer-apps): 1 case study

## Legal
- [Terms of Service](https://wouldit.work/terms)
- [Privacy Policy](https://wouldit.work/privacy)

## Optional
- [llms-full.txt](https://wouldit.work/llms-full.txt): full Markdown text of all case studies with sources
- [Sitemap](https://wouldit.work/sitemap.xml)