App Store Ratings vs Reviews: How the System Actually Works
Per-country, not global
Every storefront keeps its own rating average and count. A 4.8 in the US does nothing for your 3.9 in Germany — German users only see the German number. This is why country-level rating tables (like the ones in our reports) matter: a weak market usually means a localization or payment-method problem invisible from your home storefront.
Written reviews are also storefront-scoped, and Apple only exposes the most recent ones publicly. That recency bias cuts both ways: a bad release can dominate your visible reviews within days, and a good fix can push complaints out of sight almost as fast.
The reset button, and why to leave it alone
When you submit a new version, you can optionally reset your rating average. The temptation is obvious after a bad patch — but the reset also wipes the count, and an app showing 4.2 across 12,000 ratings converts far better than one showing 4.9 across 40. Reset only after a genuine relaunch where the old product no longer represents the new one.
The healthier lever is volume: recent ratings weigh into the average just like old ones, so a sustained flow of new positive ratings (see our guide on getting more ratings) steadily dilutes a bad period without sacrificing your count.
What to actually monitor
Track three numbers per key market: the average, the count velocity (are ratings still coming in?), and the recent-review sentiment. The average moves slowly; the recent reviews are your early-warning system. A stable 4.6 with a sudden cluster of 1-star sync complaints is a fire alarm the average won't ring for weeks.
That's the exact gap this site's reports fill: the country table shows where you stand, and the complaint themes show what's about to drag the number down.