I analyzed 50,000 tweets to find validated SaaS ideas - here's what I built

Hard to say ideas are validated... Ideas are never validated until executed. You did effectively find pains. But IMO validation (from a business POV) is really tied to execution and revenue.

I like this idea.
FYI, I tried signing up for the newsletter/emails, and the input disappears and reappears with every keystroke.

I’ve built a platform that helps businesses understand how visible they are in AI search engines like ChatGPT, Perplexity, and Claude. It shows a clear visibility score, tracks where and how a brand is mentioned, and compares performance against competitors. As AI becomes a primary discovery channel, this gives companies a way to measure and improve their presence where traditional SEO no longer applies.

Interesting approach. How do you separate signal from complaints that don’t imply willingness to pay?
Curious which filters ended up being most predictive after analyzing that volume.

This is really interesting work. What stood out to me was your point about specificity and willingness to pay mattering more than raw complaint volume. That lines up with what I have seen too.

One thing I have noticed is that many founders struggle with the next step after idea validation, which is deciding which idea is actually worth pursuing first. Scoring ideas is powerful, but translating that into realistic value or upside is often where people get stuck.

I have found that combining idea signals with simple valuation thinking helps a lot, even at a very early stage. Things like revenue potential, comparable products, and realistic buyer behavior can quickly eliminate ideas that look exciting but will never justify the effort.

Overall, great execution. The idea scoring problem you mentioned is honestly the hardest part, not the tech.

This is interesting, especially the focus on scoring rather than raw volume.
One concern: how do you validate willingness to pay beyond language signals?
In my experience, many “pain points” sound strong but never convert.
Would love to see examples where an idea led to real customer conversations or revenue.

This is interesting. One thing I’ve noticed is that many “validated ideas” still fail because the MVP isn’t tested properly before users try it.

Have you thought about prioritizing ideas where poor UX or bugs are the main source of complaints?

Hey the site is down

This is a cool idea... Hopefully there is some paywall so the best ideas are left to paid users and not the open web.

This really resonates. I’ve tried the “scroll Twitter/Reddit for ideas” approach too and it gets exhausting fast lots of noise, very little signal.

I like the insight about specificity and willingness to pay mattering more than raw complaint volume that’s something I’ve learned the hard way as well.

Curious how you currently score “willingness to pay”? Is it inferred from language or more from repetition across users?

This is a smart approach to a problem a lot of founders quietly struggle with. The shift from manually hunting for ideas to systematically extracting validated pain points is compelling, especially the emphasis on specificity and willingness to pay over raw volume. I also like that you’re pairing ideas with MVP and tech stack suggestions—it makes the output immediately actionable instead of just “interesting.” Curious to see how the scoring model evolves as more data comes in.

Really interesting approach. One thing I’d love to see: a way to track how user pain points evolve over time. For example, are complaints about a workflow recurring, or just trending for a week? That could help founders prioritize ideas with long-term demand rather than short-lived spikes.