I built ddMarketer: validated SaaS gaps from real complaints (feedback welcome)
For years, most “startup ideas” I saw were either generic or vibes-based. Build X for Y, clone Z, etc. It always felt like guessing. I saw gummysearch shut down and I saw an opportunity.
With ddMarketer, I wanted to start from evidence, not imagination. The app mines public complaints, bug reports, and feature requests across Reddit, Hacker News, Trustpilot, Stack Exchange, forums, and X. Then a two‑pass pipeline scores each one for commercial intent, confidence, and willingness‑to‑pay so weak rants never make it through.
Only the strongest opportunities get turned into “field dossiers”: each gap comes with an MVP sketch, pricing ideas, competitors, and a basic validation playbook (where to post, DM scripts, survey prompt) so you can confirm demand before you write code.
Right now there are 800+ live gaps tracked across multiple niches, refreshed weekly. I send a free “Top 10 SaaS gaps this week” dispatch, and the full searchable dataset is available on a subscription.
I’d love feedback from the IH crowd on two things:
Does this solve a real problem for you as a builder (or is there a more valuable angle I’m missing)?
What’s confusing or off about the way I explain it on the landing page?
If you’re curious, you can see this week’s Top 10 here: https://www.ddmarketer.com/top-10 and the main site here: https://www.ddmarketer.com.
Happy to answer questions about the pipeline, data sources, or editorial gate in the comments.
One thing I'd be careful with:
A complaint can sometimes look like evidence of an opportunity while actually being evidence of something else entirely.
Those can be surprisingly difficult to separate once enough signals start pointing in the same direction.
That's the part I'd be most curious about.
Hi there. thank you for your reply. what you mentioned is exactly why gummysearch died. Besides, gummysearch was reddit based only. Instead to avoid this issue we weight by source diversity + time-spread, and down-rank clusters that all trace to one event/date
That makes sense.
What I'd still be careful about is that source diversity can sometimes increase confidence that something is happening without necessarily increasing confidence about what is happening.
A complaint showing up across multiple places is obviously a stronger signal than a single thread.
I'm just not sure it automatically earns the right to be interpreted the same way.
That's the part I find interesting.
thank you i will definitely consider this as i watch data evolve in later commits.
Makes sense.
I suspect the difficult part won't be finding recurring complaints.
It'll be deciding when multiple complaints are evidence of the same underlying problem versus multiple descriptions of different problems.
That's probably where I'd spend most of my attention.