App Store reviews API for iOS apps

Apple doesn't offer a straightforward public API for pulling review text at scale — App Store Connect gives you your own app's reviews inside its dashboard, but not a general-purpose endpoint for any app, and nothing at all for a competitor's listing. Teams that want structured review data, for their own apps or anyone else's, have had to build and maintain their own scraper.

Roketfy's App Store reviews API replaces that scraper: give it an app id and it returns review title, text, rating and the app version each review was left against, as clean JSON.

The app_id is the numeric string that appears after "id" in any App Store URL — for example, 835599320 in apps.apple.com/us/app/some-app/id835599320. If you only have an app name, call the search endpoint first with that name as the keyword; it returns matching apps along with their ids, ranked the way the store itself ranks them.

The app endpoint takes the same app_id and returns the full listing metadata — title, developer, category, rating and price — useful for confirming you've matched the right app before you spend credits pulling its reviews.

The depth parameter controls how many reviews come back, from 1 up to 500, defaulting to 25 if left out. Billing is per 25 reviews, at 10 credits for the reviews endpoint — so a call for the default depth costs the same as a call for depth 1, and a deeper pull costs proportionally more.

sort_by lets you choose most_recent or most_helpful (the default), so a quick daily check can pull only the newest reviews rather than re-fetching the same top-voted ones every time.

Yes. Both the reviews and search endpoints accept a location — by name ("Istanbul,Turkey") or numeric code — and a language, by code or full name. An app's App Store listing and review pool differ by storefront, so pulling reviews for the Turkish storefront and the US storefront separately gives you two genuinely different datasets, not the same reviews translated.

This is especially useful for apps with a large non-English user base: a global rating average can mask a serious, localized problem — a broken translation, a regional payment failure — that only shows up when you pull that storefront on its own.

Review text is a direct line to the language real users apply to your app and its competitors — the features they mention wanting, the words they use to describe a problem, the comparisons they draw to other apps. Feeding that text into your own keyword research alongside the search endpoint's ranking data closes the loop between what users say and where the app actually ranks for related terms.

Pairing reviews with the app endpoint's own rating and category data also lets you track whether ASO changes — a new screenshot set, an updated description — correlate with a shift in review sentiment or volume over the following weeks.

Very large review or search pulls occasionally take longer than a single request can wait for. When that happens, the response carries a task_id rather than an error you have to interpret yourself — pass that task_id back on a follow-up call to collect the completed result instead of starting the whole pull over from scratch.