I Tried To Find A Startup Idea By Reading 11,000+ Posts From Entrepreneurs About Their Biggest Problems (i Will Not Promote)

After two years trying to find a business idea to start up something with the savings my wife and I have, we found ourselves stuck, because everything looks like a good idea, until you start it. You're judgement and motivation get clouded really quickly with the uncertainty of: "Is this something people would want?"

Instead of brainstorming startup ideas, I started reading posts where entrepreneurs talk about the problems they struggle with the most... Because that’s usually where good ideas come from.

After a while I thought: there has to be a more efficient way to do this. (duh! like this hasn't been a thing for ages LOL)

So, I started coding a small app that could read thousands of posts and identify patterns. The idea was simple: connect a few APIs, let an LLM read everything, and surface the problems entrepreneurs mention over and over.

It was supposed to be a 3-week side project.

It ended up taking 7 months.

Why? I wanted it to give me real input, and internet is actually full of junk and repetitive complaints on the same topics.

So, instead of just extracting problems, I started building an engine that turns them into a visual map of opportunities based on real pain people talk about online.

The way I think about it is like a radar.

Instead of sailing into a sea of uncertainty hoping your idea works, you can see where clusters of real problems are forming and explore the ones that match your skills, interests, or experience.

Right now, the system has analyzed thousands of posts from places where entrepreneurs talk about their challenges and clusters the most repeated problems into opportunity “bubbles.” I have my doubts that it was working, but it even started showing seasonal trends. That gave me the confident to know that this actually does what I expect it to do.

I mainly built it for myself, but I figured it might be interesting to share.

Curious what people think, is this something you’d actually use when looking for a startup idea? What other options are out there?