Two signals

App reviews and ratings come together but say different things. The rating is a single number per user. The text says what they noticed: a bug, a feature, a price, a support experience.

When they disagree

  • Four stars: "Great app, but it crashes when I upload photos."
  • One star: "Love the content, but I cannot log in after the update."
  • Rating only: no reason at all.

These are where the useful detail is.

Why text moves first

A new bug starts as a handful of complaints inside otherwise positive reviews. By the time it pulls the average down, many users have hit it. Measuring complaint share by feature area from the text shows it earlier.

Using both

  • Report the rating and the rating of new reviews by version.
  • Report complaint share by feature area from the text.
  • When the rating moves, find which areas moved.
  • When an area moves, act before the rating does.

Free review snapshot

Want this done for your own reviews?

No commitment. We use your details only to send your snapshot and follow up. Privacy policy

An example

An app's rating held steady across a release while upload-related complaints tripled in four- and five-star reviews. The team fixed uploads in the next version, before the rating moved at all.

The same for stores

Stores

Star rating vs review sentiment: why the average lags.

Star rating vs review text why the average lags

Common mistakes

  • Tracking the rating only.
  • Ignoring complaints in positive reviews.
  • Treating rating-only reviews as informative.
  • Reacting to the rating without reading reviews.

Common questions

Is the rating useless?

No; it is what users see first. It just lags and lacks reasons.

Should we prompt for ratings?

In a clear, non-deceptive way at good moments, without incentives.

Do replies affect ratings?

Not directly; users notified of fixes sometimes update.

How often should we check?

Weekly after releases, monthly otherwise.

Checklist

  • Rating with new-review rating by version.
  • Feature complaint shares from text.
  • Complaints in positive reviews captured.
  • Action before the rating moves.

In short

The rating is the verdict; the text explains it and moves first. Report the rating, decide from feature-level complaint shares in the text.

Measuring the text

Classify each review into feature areas with sentiment for each mention. For every area, track the share of reviews complaining about it, by version and month, with the count beside it. Put the rating on top of the same report; the areas underneath explain its movements.

Better rating prompts

Play's policy asks developers to prompt for ratings in a clear, non-deceptive way and prohibits incentives and forced pop-ups. Prompt at natural moments, such as after a successful order, and ask everyone the same way. A rating built from ordinary users is more accurate than one shaped by who you choose to ask.

Then read what those users write: the text is where you learn what to build and fix next.

Rating-only reviews

Many app ratings come with no text. They count towards the average but say nothing about why. Report how many ratings have text, and treat the text as a sample of what your users think: usually a large, informative sample, but not everyone.

Doing this with Falcon

Falcon classifies every Google Play review of your app into feature areas and specific issues with the sentiment of each mention, on the same taxonomy as your store reviews but kept separate. Negative share by area and month shows what is new or growing, and every issue opens the reviews behind it.