About GapTheory

Every day, people describe exactly what software fails them: in app reviews, developer forums, support threads, and complaint filings. It is the most honest market research that exists, because nobody writes an angry review to be polite. But it is scattered across millions of posts, so founders make six-month bets on a hunch and a few conversations instead.

GapTheory reads that public record continuously, clusters the recurring unmet needs into named, scored Gaps, and lets you put your own idea up against the evidence before you commit to building it.

Confidence must be earned, not asserted. We do not print a confidence score, because a number that tells you what you want to hear is worthless. Instead, every validation includes the evidence against you: people saying the problem is already solved, or that they are happy with a workaround. If your conviction survives the counter-evidence, it is real.

Evidence over opinions. Every Signal in GapTheory links back to a real post by a real person, with a source and a date. You never have to take our word for anything; click through and read the original.

We are not an idea generator. GapTheory will not tell you what to build. AI can generate a plausible guess; it cannot count real demand, and it cannot remember whether a pain was growing last quarter. We show you what people are actually asking for, why the evidence might be misleading, and you make the call.

We only collect from official APIs and endpoints that platforms publish for that purpose: Apple App Store reviews, Hacker News, GitHub, Stack Exchange, Product Hunt, and CFPB consumer complaint filings.

We do not scrape platforms that prohibit it, even where the data is publicly visible, which is why some obvious sources are missing from that list. We would rather have a smaller corpus we are allowed to have than a bigger one we are not. Evidence you might be forced to delete is not evidence you can build a company on, and that goes for us too.

I've always been an idea guy. New ways to do things, businesses to start, products that should exist. But most of my ideas died one of the same three deaths: too expensive to build, beyond my skill set at the time, or already built by a company with infinite funding.

When AI arrived, I thought that would change everything. And for the building part, it did. I code with AI every day, and GapTheory itself runs on it. But ask AI for ideas and you get the same recycled suggestions everyone else is getting, dressed up as breakthroughs. The creativity, the weird niche insight, the out-of-the-box thinking: that part is still human.

That's why GapTheory is deliberately not an idea generator. There are thousands of those. It does the opposite job: it takes the idea you already have and shows you the evidence. For it, and just as importantly, against it.

Validation used to mean weeks of cold outreach and forum-trawling, hoping strangers would respond. But the feedback already exists. Every day, thousands of people describe exactly what frustrates them, in public. GapTheory collects that from official sources, organizes it into evidence you can actually read, and lets you weigh your idea against reality.

Your idea is the creative part. GapTheory is the proof part. Get the evidence, then go build something.

Jacob Wright

@mrjacobwright

GapTheory is in private beta. The catalog currently tracks 100,000+ Signals across 2,500+ Gaps, growing daily. If you want in, join the beta; if you want to see the product first, the docs show exactly how it works.

Questions, skepticism, or a source you think we should add: hello@gaptheory.io. We read everything.