App Store & Google Play Review Monitor — reviews in one feed with alerts

Monitor app store reviews across the Apple App Store and Google Play and catch new ones the moment they land. An app review scraper / extractor / data API / monitor / feed that reads each store's public reviews, normalizes every review to one schema with a star rating, version tag, country, app name, and a rating-based sentiment label, and turns repeated runs into a new-review monitor — for product teams, ASO agencies, and competitor-review watchers.

Quick Start: click Start with the default input. With zero configuration it pulls a well-known public app's most recent App Store reviews and returns clean rows — rating, rating sentiment, title, body, app version, country, author, and date. To track your own apps, paste numeric Apple App Store app IDs into appIds (e.g. 389801252 — the number after id in an App Store URL) and/or Google Play package IDs into googlePlayAppIds (e.g. com.spotify.music). First useful results in under two minutes.

What it does

Both stores expose public reviews without a login, so the data is clean and easy to extract — no browser, no keys. This actor:

• pulls an app's most recent reviews with rating, title, body, app version, and author — from the Apple App Store (public customer-reviews JSON feed) and Google Play (the public Play Store reviews endpoint), normalized into one feed;
• enriches Apple reviews with the app name via the iTunes Lookup API (once per app);
• labels each review with a rating sentiment (positive / neutral / negative) derived purely from the star rating — not from analyzing the review text (see the note below);
• filters by rating range and keywords so you can watch, e.g., only 1–2 star reviews mentioning crash or refund;
• on a schedule, becomes a monitor — emitting (and charging for) only reviews that are new or whose text changed (e.g. an edited review) since the last run.

About ratingSentiment: it is a label derived only from the number of stars (4–5 = positive, 3 = neutral, 1–2 = negative). It is not natural-language sentiment of the review text — a 5-star review with an angry message is still labelled "positive". Use the raw rating field when you need the exact star count.

Who it's for

App product teams watching feedback per release, ASO agencies, reputation managers, and teams tracking competitor reviews across both stores.

Sources

• Apple App Store — the public, no-auth customer-reviews JSON feed (with iTunes Lookup for the app name).
• Google Play — the public Play Store reviews endpoint the Store details page itself uses to load its review list (no login, no browser). Supply one or more package IDs in googlePlayAppIds. Note: Google Play reviews have no separate title, so title is empty for Google Play rows; everything else (rating, body, version, author, date, votes) is populated.

Input

Runs zero-config (App Store demo app). Optional fields:

• appIds — numeric Apple App Store app IDs to monitor (the number after id in an App Store URL). Leave empty for a demo run on a known public app.
• googlePlayAppIds — Google Play package IDs to monitor (e.g. com.spotify.music).
• country — two-letter storefront country code (e.g. us, gb). Applies to both stores.
• countries — multiple storefront country codes (one feed per country, merged). Overrides country when set.
• minRating / maxRating — keep only reviews within a star-rating band (1–5).
• keywords — keep only reviews whose title or body contains any of these terms.
• maxItems — cap on reviews returned per run (keeps runs cheap). The budget is shared round-robin across every requested app and storefront, so each gets coverage rather than the first app draining it all. Note: Apple's public feed serves only its most recent reviews per app (~the latest few hundred), so very large maxItems values may return fewer Apple rows.
• maxCostPerRunUsd — hard ceiling on spend per run (default $5).

Output

One AppReview per row: store, appId, appName, rating, ratingSentiment, title, body, version, country, author, date, and the review sourceUrl, plus a stable store-scoped id. See .actor/dataset_schema.json.

Pricing

Pay-per-event: a small start fee plus a per-result charge, with a hard per-run cost cap on by default — so the bill is never a surprise. In monitor mode you pay only for new or changed reviews. Built on each store's public reviews endpoints, so it is low-maintenance — fixes within 24h.