App Review Triage Monitor — App Store & Google Play Alerts

Auto-triage new App Store and Google Play reviews into bugs, billing complaints, and feature requests — cluster recurring themes, catch rating drops, and get alerted in Slack. No code.

What it does

App Review Triage Monitor watches the reviews for your iOS and Android apps and sorts each new review as it lands. Instead of scrolling store listings by hand, you get reviews already bucketed by intent:

• Bugs and crashes — reviews reporting freezes, crashes, login failures, and broken features.
• Billing complaints — refund requests, accidental charges, subscription and pricing gripes.
• Feature requests — "I wish it could…" reviews that signal what users want next.

On top of the buckets, it:

• Clusters recurring themes so you see "17 reviews this week mention the checkout crash" instead of 17 separate one-star reviews.
• Flags rating drops when your average rating slides, so a regression in a new build doesn't go unnoticed.
• Tracks competitor apps side by side, so you can compare what users complain about across rivals.
• Sends a webhook (e.g. to Slack) when new reviews come in, and can produce a weekly report summarizing what changed.

Triage and clustering are heuristics to help you read faster — not ground-truth sentiment scoring. You can run pure keyword/rule triage for free, or pass your own Anthropic API key to enable LLM-assisted triage for messier, multi-language reviews.

Who it's for

Mobile product managers, app founders, and support / CX teams who watch App Store and Google Play reviews and currently read them by hand — and miss the signal in the noise. If review reading is a manual Monday-morning chore and bug reports hide between five-star ratings, this is built for you.

Quick start

You can run this Actor two ways: as a monitor (scheduled or one-off runs on the Apify platform) or as an MCP tool your AI assistant calls directly.

Run the monitor

1. Open the Actor on Apify and click Try for free.
2. In Input, add one or more apps to watch (see Input below).
3. Click Start. Results appear in the run's dataset; if you set webhookUrl, new reviews are pushed there too.
4. To run it on a schedule, open Schedules in Apify and point a new schedule at this Actor.

Minimal input:

{
  "apps": [
    "https://apps.apple.com/us/app/instagram/id389801252",
    "com.instagram.android"
  ],
  "country": "us",
  "maxReviewsPerRun": 200,
  "onlyNewSinceLastRun": true,
  "detectBugsCrashes": true,
  "detectBillingComplaints": true,
  "detectFeatureRequests": true,
  "detectRatingDrops": true,
  "clusterThemes": true
}

Connect it to Claude, Cursor, or another MCP client

The Actor is exposed through Apify's hosted MCP server over streamable HTTP. The endpoint is:

https://mcp.apify.com/?actors=dami_studio/app-review-triage-monitor

Authenticate with a Bearer APIFY_TOKEN or with OAuth in the browser.

Claude Desktop / Cursor (remote connector): add a custom MCP connector and paste the URL above. You'll complete OAuth in the browser, so no token sits in a config file.

JSON config (clients that bridge stdio to remote, e.g. via mcp-remote):

{
  "mcpServers": {
    "app-review-triage": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://mcp.apify.com/?actors=dami_studio/app-review-triage-monitor",
        "--header",
        "Authorization: Bearer ${APIFY_TOKEN}"
      ]
    }
  }
}

Set APIFY_TOKEN to your token from Apify → Settings → API & Integrations. Once connected, ask your assistant something like "Triage this week's new reviews for my app and group the bug reports."

Input

Field	What it does
apps	The apps to monitor — App Store URLs/IDs and Google Play package names.
competitorApps	Competitor apps to track alongside yours for side-by-side comparison.
maxReviewsPerRun	Cap on how many reviews to pull per run.
onlyNewSinceLastRun	Only process reviews that appeared since the previous run.
country	Store country/region to read reviews from (e.g. us, gb, de).
detectBugsCrashes	Bucket reviews reporting bugs and crashes.
detectBillingComplaints	Bucket refund, charge, and subscription complaints.
detectFeatureRequests	Bucket reviews asking for new features.
detectRatingDrops	Flag when the average rating drops.
clusterThemes	Group reviews into recurring themes.
weeklyReport	Produce a weekly summary of what changed.
webhookUrl	URL (e.g. Slack incoming webhook) to push new reviews to.
useLLMTriage	Use LLM-assisted triage instead of rule-based only.
anthropicApiKey	Your Anthropic API key, required when useLLMTriage is on.

Output

Each result row represents one triaged review:

{
  "store": "app_store",
  "appId": "id389801252",
  "country": "us",
  "reviewId": "1099…",
  "rating": 1,
  "title": "Crashes on launch",
  "text": "Latest update crashes every time I open the camera tab.",
  "author": "user1234",
  "date": "2026-06-21T09:14:00Z",
  "categories": ["bug_crash"],
  "theme": "camera-tab-crash",
  "isNew": true
}

Reviews can carry more than one category. Theme labels are shared across reviews in the same cluster, so you can count and sort by theme. When rating-drop detection fires or a weekly report is generated, those summaries are written to the dataset (and/or your webhook) alongside the review rows.

Example use cases

• Triage Instagram crash reviews into a bug list
• Surface Spotify billing and refund complaints
• Pull Notion feature requests from store reviews
• Compare TikTok vs Instagram and Snapchat reviews

Pricing

$0.40 per 1,000 reviews processed ($0.0004 each) — the main charge, and the only one that scales with volume. Four smaller events cover the rest of the work:

Charge event	Free plan	Bronze and above
Review processed	$0.0004	$0.0004
Run start	$0.001	$0.001
App checked	$0.004	$0.002
AI triage generated	$0.0005	$0.0003
Report generated	$0.005	$0.003

Five charge events in total. A scheduled run over two apps that finds 150 new reviews and runs rule-based triage costs $0.001 (start) + $0.008 (two apps checked) + $0.06 (150 reviews) = $0.069 on a free plan. AI triage and weekly reports are only billed when you switch them on.

If you enable LLM-assisted triage with your own Anthropic API key, the Anthropic model usage is billed separately by Anthropic, on top of the AI-triage event above.

FAQ

Does it cover both the App Store and Google Play? Yes. Add Apple App Store apps and Google Play apps in the same run. Store review availability varies by app and country, so coverage depends on what each store exposes for a given listing and region.

Is the bug / billing / feature sorting accurate? It's a heuristic to help you read faster, not ground-truth sentiment analysis. Rule-based triage catches the common patterns; turning on LLM-assisted triage (with your Anthropic key) handles messier and multi-language reviews better. Either way, treat the buckets as a fast first pass, not a verdict.

Can I get a Slack alert when new bad reviews come in? Yes. Set webhookUrl to a Slack incoming webhook (or any HTTP endpoint) and run on a schedule. New reviews are pushed there as they're found.

Do I need to write any code? No. Fill in the input form and run it, or schedule it. The MCP integration also lets an AI assistant call it for you with no setup beyond connecting the server.

How do I monitor competitors' app reviews too? Add their App Store URLs/IDs and Google Play package names to competitorApps. You'll get their reviews triaged the same way, so you can compare complaint themes across apps.