App Store Connect opens on your home storefront, so that is the rating you know. It is one of many, and on a distributed app it is frequently not the important one.
Apple is unambiguous about this. From Ratings and reviews overview:
"Your App Store product page displays a single overview rating for your app for each country or region of the App Store"
And on the mechanism:
"The country or region associated with a rating reflects the location where the customer initially purchased the app."
Apple's ratings and reviews page says the same from the other direction: "This summary rating is specific to each territory on the App Store."
Apple currently publishes 175 as the number of App Store countries or regions — though it names them differently on different pages, calling them "175 storefronts" on apple.com and "175 countries or regions" in App Store Connect Help. Whichever noun you prefer, you have up to 175 ratings, not one.
| Question | Apple's answer | Where |
|---|---|---|
| Is the rating per country? | Yes — one overview rating per country or region | Ratings and reviews overview |
| What decides a rating's country? | Where the customer initially purchased the app | Ratings and reviews overview |
| Can I reset one country only? | No — a reset affects all regions simultaneously | Reset an app overview rating |
| Do reviews reset too? | No — written reviews continue to display | Reset an app overview rating |
| Do ratings affect search ranking? | Yes, they "can influence" it | App Store search |
| Is that influence per storefront or global? | Not stated anywhere | — |
| Minimum ratings before an average shows? | "Enough submissions" — no number published | — |
What a sweep actually finds
Because the public lookup endpoint returns rating count and average per storefront, this is measurable without any account access. Appstro's worldwide sweep queries 58 storefronts and reports each one's count and average.
Two findings from running it:
The scale gap is enormous. One large app read roughly 8.9 million ratings in the US storefront and 1.4 million in Germany — same app, same moment, two entirely separate numbers.
Your average is not the average you think. The sweep combines storefronts into a ratings-weighted figure, never a mean of the per-storefront means. A storefront with 40 ratings and a storefront with 400,000 cannot count equally, and a naive average of averages lets the quiet ones swing the result.
Against that weighted figure, a storefront is flagged as weak when it sits at least 0.2 stars below it on 20 ratings or more. The 20-rating floor is doing real work: without it, one unhappy user in a small storefront produces a 1.0-star average and tops the list of problems.
A storefront that fails to answer is reported as unavailable, never as zero. One dead lookup must not be read as a market that does not rate you.
The reset is global, and Apple's pages disagree about it
This is the trap worth knowing before you touch anything.
Apple's marketing page describes the summary rating as specific to each territory and mentions that you can reset it — which reads as though the reset is per-territory too. The Help page is explicit that it is not. From Reset an app overview rating:
"When you reset your app's overview rating, it affects all regions or countries simultaneously."
Two further constraints from the same page: a reset "only applies to the platform version that you reset your rating for", and "it won't apply to written customer reviews, which will continue to display".
So you cannot clear a bad rating in one market and keep a good one in another. Follow the Help page — it is the operational one, and it is unambiguous.
The second question the sweep raises
Once you know where your ratings come from, a different question appears: are those users reading their own language?
Rating counts and listing text come back in the same response, so both halves cost one set of requests rather than two. Comparing each storefront's name and description against the US listing gives you the storefronts where people already found you, already rated you, and are reading English.
Verified against live data: WhatsApp in India carries 7,996,710 ratings against a description byte-identical to the US listing — 2,382 characters, same on both. Across a full 58-storefront run, 40% of that app's worldwide ratings came from storefronts reading the English listing.
That is the ranking Appstro's localization priority produces, and it is deliberately narrow about what it claims. These users found the app and rated it. Whether translating the listing would produce more of them is not something a public dataset can answer.
Two details from building it, both of which change the answer:
- The localized test is an OR across name and description, not name alone. WhatsApp keeps an identical product name in every storefront while returning localized descriptions; testing the name only would have called every storefront unlocalized.
- The US baseline is fetched with
lookup, never taken from a search result, because the two endpoints disagree. Observed at the same moment for the same app: 144,886 ratings from search and 144,851 from lookup. Small, but it would mean measuring one storefront off a different endpoint from the other 57, on precisely the number the ranking sorts by.
What Apple does not say, and it matters
Apple documents that ratings and reviews can influence App Store search ranking. Apple says nothing about whether that influence is scoped to the storefront the ratings came from or pooled across all of them.
Both of the natural conclusions — "a bad rating in Brazil drags my US ranking" and "a bad rating in Brazil only affects Brazil" — are unsupported. This was checked directly and there is no Apple statement in either direction.
What is verifiable, and quite sufficient to act on, is that users in that storefront see that rating, and it is the number they decide on.
What this cannot tell you
- Not whether a low rating in one country affects any other. Apple publishes nothing; nobody outside Apple knows.
- Not the minimum ratings before Apple displays an average. Apple says only "if an app has enough submissions in your country or region" — no threshold is published, so no threshold is printed here.
- Not how the star average is computed. No aggregation formula, weighting or time decay is documented anywhere.
- Not why a storefront rates you badly. A count is not a cause. The review text for that storefront is, and it is one drill-down away.
- Not per-country downloads. Ratings are per-storefront in Apple's public data; downloads are not.
The work
Run the worldwide sweep once and look at the shape of it — most developers find at least one storefront in their top five that they had never thought about. Then run localization priority over the same app to see which of those storefronts is reading a language it did not choose.