Concept

Where PainHunt's data comes from, source by source

The PainHunt Team · May 29, 2026 · 3 min read

TL;DR: PainHunt's database holds pain points from 26 platforms. Eighteen are collecting right now, four are stalled on faults we're fixing, and four are switched off. This page names all of them, including the broken ones, because a data product that won't tell you where its gaps are isn't worth much.

Why source diversity matters

Any single platform has a personality. App Store reviews are short and emotional. GitHub issues are precise and technical. Hacker News skews toward developer and founder problems. If you only listen to one, you inherit its blind spots.

PainHunt deliberately mixes source types so a problem that's invisible on one platform still gets caught on another. The goal isn't raw volume — it's coverage of how real demand actually expresses itself.

Collecting right now

Social & forum signal

  • Hacker News (stories + "Who is hiring")
  • Lemmy (federated communities)
  • Mastodon
  • Bluesky
  • 15 SaaS-focused Discourse communities

App stores (many countries)

  • Apple App Store reviews (19 countries)
  • Google Play reviews (19 countries)

Developer & technical communities

  • Stack Exchange network

Discovery & launch platforms

  • Product Hunt
  • BetaList
  • Chrome Web Store

Writing & long-form

  • Medium
  • Substack
  • Dev.to

Jobs & remote work

  • Remotive
  • WeWorkRemotely

Video

  • YouTube

Stalled right now

These have data in the database and are meant to be collecting, but aren't at the moment. Each has a specific, boring cause:

Source Stalled since Cause
GitHub Issues ~5 days API credential expired; needs reissuing
GitHub Discussions ~5 days Same credential
RemoteOK ~10 days TLS handshake failures through our egress path
AppSumo ~16 days The listing page we crawled was removed in a site redesign

We'd rather list these than quietly count them as active. Everything they collected before the fault is still in the database and still searchable.

Switched off

Source Reason
Twitter / X Access restrictions made reliable, compliant collection impractical
Reddit Datacenter-IP blocking prevented stable collection despite rate-limit compliance
Hashnode The platform moved its API behind a paywall
Telegram Low hit rate and heavy overlap with Product Hunt — not worth the crawl budget

One more worth naming: TrustMRR was evaluated as a competitor-revenue source and never shipped. Its terms don't permit the redistribution the feature would have needed, so it collected nothing. If you've seen it mentioned in older material, that's why it isn't in the list above.

The mix changes as platforms change their access policies. What matters is breadth across source types, not any single site — and the count is deliberately not baked into this page's title, because it would be wrong within a month.

How the data is handled

  • Summaries, not reprints. PainHunt extracts and scores pain points and links back to the public original. It does not republish the author's full text.
  • English-facing. The user-facing product presents English content; internal analysis fields are never exposed.
  • Continuous, not static. Crawlers run on an ongoing basis, so the database grows daily.

What this means for you

When a pain point ranks highly in PainHunt, it's usually because the same kind of problem appears across multiple source types — an App Store review and a forum thread and a Stack Exchange question. That cross-platform agreement is a stronger signal than any single loud complaint.

It also means you should read a platform filter as "where this surfaced", not "where the problem exists". A billing complaint that only shows up in app store reviews is still a billing complaint; it just means the people hitting it are consumers rather than developers.

To see it in practice, open the Pain Point Browser and filter by platform, or read how PainHunt works for the scoring details.

Frequently asked questions

How many data sources does PainHunt use?

The database holds data from 26 platforms. Eighteen of them are collecting continuously right now; four are temporarily stalled on technical faults we're working through; four are switched off. We publish the current split rather than a single headline number, because the number moves.

Does PainHunt use Twitter or Reddit?

Not currently. Both are switched off — Twitter since its access changes, Reddit because datacenter-IP blocking prevented stable collection even while we stayed inside its rate limits. Their historical data is still in the database and still searchable; it just isn't growing.

How does PainHunt handle the original content?

PainHunt stores and displays AI-generated summaries and scores, and links back to the public source. It does not republish the original author's full text.

Validate your idea against real demand

PainHunt scores hundreds of thousands of real user complaints by commercial potential — so you build what people already want.

Open the Pain Point Browser

Keep reading

Where PainHunt's data comes from, source by source | PainHunt