App Review Intelligence Agent

What does App Review Intelligence Agent do?

App Review Intelligence Agent turns public reviews from the Apple App Store and Google Play into product intelligence, not just a review export. It fetches recent reviews, normalizes them, scores sentiment and urgency, tags recurring complaint categories, and writes an executive report for product, founder, ASO, marketing, and agency workflows.

Use it as an app review analysis API when you need answers such as "what is making users churn?", "which missing features keep coming up?", or "what should we fix before the next launch?"

Want the finished report instead of running the Actor?

I also offer a fixed-scope App Review Intelligence Audit: one iOS App Store or Google Play app, up to 500 recent public reviews, complaint clusters, churn risks, feature requests, ASO opportunities, and prioritized product actions. It costs €150 one-time and is delivered within 48 hours.

Use the Actor below when you want automation and API access. Use the audit when you want a report ready to review or send to a client.

Why use App Review Intelligence Agent?

Raw app review scrapers are useful, but they leave the hard part to you. This Actor adds a product-analysis layer on top of review collection:

• Normalized iOS and Google Play review rows in one dataset.
• Sentiment and urgency scoring before any LLM is called.
• Keyword-based categories such as bug, crash, pricing, subscription, onboarding, support, ads, performance, UX, feature requests, privacy, auth, map/navigation UX, content coverage, and competitor/value objections.
• Complaint clusters with counts, severity, and example quotes.
• Churn risks, ASO opportunities, support opportunities, and prioritized product actions.
• Optional OpenRouter synthesis using deepseek/deepseek-v4-flash by default when OPENROUTER_API_KEY is available.

Because it runs on Apify, you can schedule it, call it through the API, monitor runs, and export results to JSON, CSV, Excel, HTML, or integrations.

Live sample

A live Pokémon GO Google Play run analyzed 12 recent reviews in 24 seconds and produced an LLM-enriched report with churn risks, performance complaints, account-linking friction, ASO opportunities, and prioritized product actions.

• Run ID: qJWJUlOsbQbiM105t
• Dataset ID: lkQNT0XaP45hFgTfM
• Done-for-you audit: https://rgo.pt/services/app-review-audit

What data can App Review Intelligence Agent extract?

Field	Type	Description
type	string	review for normalized review rows or intelligence_report for report rows
source	string	appStore or googlePlay
rating, title, text, author, version, date	mixed	Public review details from the store
sentiment, sentimentScore, urgencyScore	mixed	Deterministic scoring fields
categories, signals	array	Complaint categories and short extracted phrase labels
topComplaintClusters	array	Report-level clusters with count, severity, and examples
featureRequests, churnRisks, asoOpportunities	array	Product and marketing intelligence outputs
markdownReport	string	Human-readable executive report

How to analyze App Store and Google Play reviews

1. Add one or more Apple App Store IDs or URLs, Google Play package names, or Google Play URLs.
2. Choose sources, countries, languages, review limit, and sort order.
3. Optionally add a focus such as "Find churn risks, missing features, pricing objections, and onboarding complaints."
4. Add OPENROUTER_API_KEY as an Apify secret if you want LLM-enriched report writing.
5. Run the Actor and open the dataset. Review rows and final report rows are written to the same default dataset.

How much will it cost to analyze app reviews?

This Actor is designed for pay-per-event pricing. It charges for analysis start, each review processed, and each intelligence report generated. In practice, cost scales with maxReviewsPerApp, number of apps, countries, languages, and selected sources.

Current scheduled pricing, configured through Apify API and subject to Apify's 14-day notice rule for switching a free Actor to paid:

Event	Price	Meaning
analysis-start	$0.00005	One real public-store analysis run starts
review-processed	$0.002	One public app review is fetched, normalized, scored, and analyzed
intelligence-report	$0.05	One product intelligence report is generated

Example: 100 reviews + one report costs about $0.25005 before platform-usage/free-plan nuance. Set a small review limit first when testing a new app, then increase it for ongoing monitoring or competitor research.

Input

See the input tab for full configuration options.

Example input:

{
  "sources": ["appStore", "googlePlay"],
  "appStoreAppIds": ["284882215"],
  "googlePlayAppIds": ["com.whatsapp"],
  "countries": ["us"],
  "languages": ["en"],
  "maxReviewsPerApp": 50,
  "sort": "newest",
  "includeLlmReport": true,
  "model": "deepseek/deepseek-v4-flash",
  "focus": "Find churn risks, missing features, pricing objections, and onboarding complaints.",
  "dryRun": false
}

dryRun uses built-in fixture reviews and does not fetch public stores. It is useful for checking output shape.

Output

You can download the dataset in various formats such as JSON, HTML, CSV, or Excel.

Simplified output sample:

[
  {
    "type": "review",
    "source": "googlePlay",
    "appId": "com.example.app",
    "appName": "Example App",
    "rating": 2,
    "text": "The app crashes after login and support has not replied.",
    "sentiment": "negative",
    "sentimentScore": -0.75,
    "urgencyScore": 0.79,
    "categories": ["crash", "support"],
    "signals": ["crash: crash", "support: support"]
  },
  {
    "type": "intelligence_report",
    "appId": "com.example.app",
    "appName": "Example App",
    "reviewCount": 50,
    "averageRating": 3.4,
    "negativeRate": 0.32,
    "topComplaintClusters": [
      {
        "topic": "crash",
        "count": 8,
        "severity": 0.74,
        "exampleQuotes": ["The app crashes after login and support has not replied."]
      }
    ],
    "executiveSummary": "Moderate negative review pressure for Example App...",
    "markdownReport": "## Example App review intelligence\n..."
  }
]

Data source note

Review fetching uses public Apple App Store and Google Play review endpoints through the app-store-scraper and google-play-scraper npm packages. No store credentials are required. Store availability, localization, ranking, and returned review counts can vary by country, language, and source behavior.

OpenRouter and LLM note

The Actor always runs deterministic analysis first. If OPENROUTER_API_KEY is available and includeLlmReport is true, it asks OpenRouter to improve the report using the selected model, defaulting to deepseek/deepseek-v4-flash. If the key is missing or the LLM call fails, the Actor still outputs normalized reviews, deterministic scoring, and a markdown report.

The API key is read from the environment and is never printed or written to the dataset.

FAQ, disclaimers, and support

Is this just a review scraper?

No. It does fetch public reviews, but the output is designed around decisions: clusters, risks, product actions, support opportunities, and ASO messaging. The normalized review rows are included so teams can audit the report.

Can I use it for competitor research?

Yes. Add competitor App Store IDs or Google Play package names and compare complaint clusters, feature requests, review sentiment, and ASO opportunities.

Is personal data included?

Our Actors are ethical and do not extract any private user data, such as email addresses, gender, or location. They only extract what the user has chosen to share publicly. We therefore believe that our Actors, when used for ethical purposes by Apify users, are safe. However, you should be aware that your results could contain personal data. Personal data is protected by the GDPR in the European Union and by other regulations around the world. You should not scrape personal data unless you have a legitimate reason to do so. If you're unsure whether your reason is legitimate, consult your lawyers.

Use the Issues tab for feedback and the API tab for programmatic access.