How to find a winning SaaS idea

Hey! 👋 Here's what I promised you:

The big mistake many people fall into: they come up with a SaaS idea they like, without checking whether the market has the same problem — and they waste entire months on nothing. The right way: let Claude search for the real problems first, then validate before you build.

How to start

1. Pick a very narrow niche — not just "ecom", but something more niched like dentists, stores, case management...
2. Copy this prompt into Claude:

Act as a SaaS market researcher. My niche: [DENTISTS / ECOM BRANDS / STORES...].
Find the 10 most painful, recurring problems this niche complains about online
(Reddit, forums, reviews, blogs). For each problem give me:
- who exactly has it
- how they solve it today and why that solution sucks
- a super-niched SaaS idea that fixes it
- pain level /10 and willingness to pay /10
Rank them and tell me which ONE you would build first and why.

3. Pick the idea that stood out to you, and go search for it on Google: collect the links of the blogs and subreddits where people complain about the same problem.
4. Go back to Claude, share the links with it using this second prompt:

Here are links where people complain about this problem: [PASTE LINKS].
Analyze them all and generate a full report: is this pain real and frequent?
Who exactly has it? Would they pay for a fix?
Give me a final verdict /10, the risks, and the best angle for a niched SaaS.

5. If the verdict is high — you've got a validated idea, not just hype. Now you can start building. ✅