I built a tool that finds startup ideas from real complaints. 4 months in, here's what changed.

Most founders build first, validate never. I did that twice. Third time I flipped the order.

PainBase started as a personal script: crawl Reddit, HN, Dev.to, X, Lobsters and Indie Hackers, pull complaints, score them by severity vs how many solutions already exist. If people are angry and nothing solves it, that's a gap.

I turned the script into a product. 4 months later, here's what it does now:

- AI Pain Point Reports: pick a niche, get a structured report with source-backed complaint links (no hallucinated "trends", every pain point links to the actual post)

- Real-time pain point feed so you catch complaints as they happen

- Competitor deep analysis: what users hate about existing tools in your space

- One-click MVP landing page builder (HTML export) so you can test demand same day

- New: real-time lead gen with Telegram alerts. It scores buying intent from live community posts and pings you before those leads get discovered elsewhere

What I learned building this:

1. Severity beats volume. 50 mild complaints < 5 people saying "I would pay to make this stop"

2. Solution availability is the filter everyone skips. High pain + crowded market = still a bad idea

3. The complaint IS the marketing copy. Users literally write your landing page headline for you

Pricing is a one-time Founder Pass at $29 (goes to $49 after launch), includes 25 AI research credits refilled monthly. No subscription.

Try it: painbase.space