Ratings against reach, not in isolation

Rating Benchmark

Four thousand ratings is excellent for a small app and dismal for a big one, so a rating count on its own answers nothing. This puts an app’s ratings next to its reach, and both next to the 25 apps it shares a search result with.

What this answers

“Should I add a review prompt” is unanswerable from a rating count, because the number that matters is ratings relative to reach, and reach is the part you cannot see for anyone else. This pairs Apple’s public rating counts with third-party download estimates for the whole cohort, so the comparison is at least consistent.

Already know the answer is yes? Worldwide reviews shows which storefronts your ratings actually come from.

Is 4,000 ratings good? It depends entirely on your reach

A rating count on its own answers nothing. Four thousand ratings is excellent for a small app and dismal for a big one. This ranks your ratings against your estimated reach, and then ranks that figure against the 25 apps sharing your search result.

It is the number behind the question people actually ask, which is whether it is time to add a review prompt.

How it works

  1. 1One search call builds the cohort, one lookup gets your app, and a single batched call gets download estimates for all of them at once.
  2. 2Each app’s lifetime rating count is divided by its download estimate to produce an index.
  3. 3The cohort is ranked and your position in it is reported as a percentile.
  4. 4Average star rating is shown alongside, because that is the one comparison here with no unit problem at all.

What it cannot tell you

  • The two numbers do not share a unit, and that is the one way this page could mislead. Rating counts are lifetime and per storefront; the download estimate is worldwide and a rolling 30 days. Their ratio is NOT a ratings-per-install rate and must never be read as a percentage of users who rate you.
  • It is an ordinal index. Every app in the cohort is computed from the same two mismatched sources, so position is informative even though the raw number is not.
  • Apps whose estimate comes back as a floor rather than a measurement are excluded and counted, because dividing by a placeholder invents a denominator — one real app with 91,101 ratings would have scored 170x its true value.
  • Age is a direct confounder: a lifetime numerator over a one-month denominator flatters an old app with fading downloads, so a warning fires when the age gap is large.

Frequently asked questions

How many ratings should my app have?

There is no absolute number. What is comparable is ratings relative to reach, measured the same way for every app in your search result — which is what the percentile on this page reports.

Should I add a review prompt?

This page will not tell you to. Where and when an app asks is a product decision it cannot see, and a prompt at the wrong moment costs you a rating and a user.

Why are some apps excluded from the comparison?

Because their download estimate came back as a floor — a "fewer than" figure rather than a measurement. Treating that as a number would fabricate the denominator, so those apps are excluded and the count is shown.