Data sources

Built from public data.

userken audiences are assembled from sources anyone can go and read. That is the whole point: it is why you can build an audience for a competitor, and why you can check our work. Here is what is live, what each source contributes, and what is next.

Coverage

Live from the index.

514,617
Reviews indexed
513,293
Vector embeddings
74
Apps covered
11
Categories
Live sources

Connected today.

Apple App Store

Live

Long form, opinionated, and heavy on the moment something broke. Strongest signal for feature complaints, pricing reactions and version regressions. Carries rating, version, country and date, which is what makes time split hold outs possible.

191,385
reviews indexed

Google Play

Live

Higher volume and shorter on average, with a wider device and market spread. Balances the Apple skew and fills in the lower end of the rating distribution, where the most specific complaints live.

323,168
reviews indexed
Next

Connectors in the queue.

Reddit

Planned

Discussion rather than verdicts: comparisons, workarounds, and the reasons behind a switch. Adds the deliberation that a store review has already skipped past.

not connected yet

Hacker News

Planned

A technical, early adopter slice with strong views on privacy, performance and pricing models. Useful as its own segment, and openly unrepresentative of a mass market.

not connected yet

Bluesky

Planned

Short, fast reactions with timestamps, which is what you want for tracking how sentiment moves in the days after a launch.

not connected yet

Your own CSV uploads

Planned

Support tickets, survey verbatims, interview transcripts, NPS comments. Uploaded data stays private to your workspace and can be mixed with public sources or used on its own.

not connected yet
How we treat it

Rules we hold ourselves to.

  • Public data only, collected from public endpoints, used to build aggregate segments rather than to profile individuals.
  • Quotes are shown as evidence with their review id, so any claim can be traced back to its source text.
  • Uploaded data stays private to the workspace that uploaded it and is never mixed into public audiences.
  • Reviewers are self selected, which we state plainly on the accuracy page rather than burying in a footnote.

Want a category we do not cover?

The pipeline is the same for any consumer category: index the reviews, embed, cluster, expose. Tell us which one you need.