1,964 founders swiped on 25,908 real problems. Almost nobody said no because the market was too small.

Disclosure first, so you know what kind of post this is: I run BigIdeasDB, a database of complaints pulled from G2, Capterra, Upwork, Reddit and app store reviews. Around 680 people pay for it and it does a bit under $9k a month. Everything below comes from our own logs, not a survey.

Back in January we shipped a swipe mode. You get a deck of cards, each one a real complaint or gap someone wrote about software they use. Swipe right if you'd build it, left if you wouldn't. When you swipe left we ask why, with seven options. It's optional, and people skip the question about a third of the time.

Since then 1,964 founders have swiped 25,908 times. 81% of swipes were a no.

Why they said no. 13,946 of those left swipes came with a reason:

[chart: generated-images/ih-2026-09-26-skip-reasons.png]

• Doesn't excite me: 36%
• Not my niche: 19%
• Too vague: 15%
• Too many competitors: 11%
• Too technical for me: 8%
• Too ambitious: 6%
• Market too small: 4%

"Market too small" came dead last. That's the reason every validation guide tells you to check first. People used it 565 times out of nearly 14,000.

"Doesn't excite me" was used almost nine times as often. More than a third of the time, a founder looked at a problem real people are paying to have solved and passed because it was boring.

What "boring" looked like. These are the cards people skipped most often for not being exciting: contract lifecycle management, voting management, disclosure management, employee scheduling, 401(k) admin, bakery software.

Every one of those is a category businesses already pay for.

The part I didn't expect. The cards that got the most right swipes weren't more exciting. They were just as dull:

• Customizable reporting for accounting software: 30 of 83 swiped right (36%)
• Automated data validation for lead generation: 40 of 112 (36%)
• An integration layer between billing and payments tools: 33 of 100 (33%)

Our overall right-swipe rate is 19%, so those did almost twice as well.

The difference, as far as I can tell, is that the losers were category names and the winners named one specific broken thing. "Contract lifecycle management" is a market. "I can't get the report my accountant needs out of the software I pay for" is a problem. Same level of boring. One of them tells you what to build on Monday.

The "too vague" numbers point the same way. Product Hunt cards, which describe products instead of complaints, got called too vague 29% of the time. Cards built from G2 reviews got that 12% of the time.

What didn't work on our side. There's a save button on every card. Across 1,964 people and eight months, it has been used 53 times. That's a product failure and it's mine. It also means I can't tell you how many of those 4,988 right swipes turned into anything anyone actually built. I only know what people said yes to, not what they did next.

One more caveat: the skip reason is what people picked from a list in about a second. It's the stated reason, not necessarily the real one. "Doesn't excite me" might be doing a lot of work for "I don't understand this industry."

One number I can't explain. Swipes made in under 3 seconds were right swipes 37% of the time. Swipes that took 3 to 10 seconds were right swipes 11% of the time. My guess is that snap yeses are gut reactions to something familiar and the longer looks are people talking themselves out of it, but I really don't know.

So, two questions:

If you've built something in a category you'd call boring, did the boredom ever actually matter once it was making money?

And for anyone who looks at data like this: what would you read into the under-3-seconds number?