One thing I’ve noticed talking to early-stage founders is how many people get stuck before they even start building. The “idea phase” turns into weeks of overthinking markets, trends, and whether something is original enough.

Recently I came across a different approach while searching on Google, instead of brainstorming from scratch, some founders are starting with curated collections of startup ideas and real-world problems.

One example I found was a database called StartupIdeasDB that compiles thousands of startup concepts gathered from places like Reddit discussions, industry trends, and product launches.

What stood out to me is that it treats ideas almost like a dataset you can analyze rather than random inspiration. Instead of staring at a blank page, you start by studying patterns of problems people repeatedly mention.

It seems to help people move from “thinking about ideas” to actually picking one and testing it. Obviously execution still matters far more than the idea itself, but reducing that initial paralysis seems useful.

Curious if others here have tried a similar approach or still prefer starting from a completely blank slate.