Google Play Review Change Monitor and Alerts

Monitor review changes, not another static export

This independent product is not affiliated with, sponsored by, or endorsed by Google or Google Play. Google Play listings and developer responses remain the authoritative public source.

This is a limited-permission Actor because its bounded named store retains only public review IDs needed for deduplication.

Google Play Review Change Monitor and Alerts watches a bounded list of Android apps and emits only newly observed reviews that match a star-rating ceiling. It is designed for recurring product-feedback, support, release-quality, and competitor-monitoring workflows where downloading the same review corpus every day creates duplicate records and unnecessary cost.

The default free preview checks the live source and returns counts, per-app source health, and the number of matching reviews that are new to the selected monitor state. Preview mode does not emit new_app_review_alert events and does not advance deduplication state. This lets a user verify app IDs, coverage, and expected volume before enabling paid delivery.

What each alert contains

Every delivered row includes the stable review ID, Android package ID, storefront country, requested language, rating, public review text, reviewer display name, publication timestamp, app version, helpful-vote count, developer reply and reply timestamp, and the public Google Play listing URL. Sentiment is derived transparently from the rating: one or two stars is negative, three is mixed, and four or five is positive.

The Actor also adds deterministic keyword labels for common operational themes: crashes and bugs, billing and subscriptions, login and account trouble, performance, advertising, and feature requests. These labels are intentionally simple and auditable. They do not pretend to be an opaque AI diagnosis and add no model charge.

Stateful monitoring

Set a stable monitorId for each independent workflow. A named key-value store retains delivered public review IDs, so later scheduled runs suppress duplicates. Different monitor IDs can watch the same apps with different rating limits or regions without sharing state. The retained state is bounded to 20,000 IDs.

State advances only for records that the platform confirms were charged and delivered. If a user-defined charge ceiling stops a run, undelivered reviews are not marked seen and remain eligible on the next run. The Actor therefore avoids silently losing alerts at a billing boundary.

Inputs and practical workflows

Provide one to twenty package IDs, for example com.google.android.apps.translate. Select a two-letter country and language, fetch between one and 1,000 newest reviews per app, and choose a maximum rating. The common production setting is three stars, which captures negative and mixed feedback without charging for routine positive reviews.

Useful recurring tasks include monitoring a new app release for crashes, tracking whether support has replied to one-star feedback, detecting subscription complaints, comparing competitor pain points, or forwarding fresh review rows to Slack, n8n, Zapier, a database, or a product analytics pipeline through Apify schedules and webhooks.

Reliability and limitations

Each requested app receives an explicit ok or error source status in preview telemetry. A partial failure is visible, and a run fails closed if every app source fails. Inputs and output volume are bounded to protect runtime and cost. The source is public Google Play data; deleted reviews, regional differences, ranking behavior, and source-side pagination can affect what is available.

This version deliberately does not advertise Apple App Store coverage. Apple's unauthenticated public customer-review feed was returning empty results during the implementation audit, including for high-volume apps. Apple support will be added only after a source can be validated without silently representing an empty or incomplete feed as complete data.

Pricing

The Actor uses Pay per event + usage with no start fee. Preview mode is free of custom events. Paid runs charge $0.005 for each delivered new_app_review_alert; platform usage is passed through separately to the runner. Users can cap the maximum charge for a run, and the delivery-safe state logic honors that boundary.