App Review Radar — Multi-Country App Store Reviews + Actionable Summary

Stop reading thousands of App Store reviews across countries.

App Review Radar collects public iOS App Store reviews from multiple storefronts, normalizes them into one dataset, and adds a one-page report of what users complain about, what they praise, and what to fix first.

No login. No proxy. No browser. Public Apple data only.

The problem

• Your reviews are spread across US, UK, Japan, Germany and more.
• The 1-star reviews are buried in noise.
• You can see the rating, but not why it dropped.
• Raw CSV exports don't tell you what to fix next.

What you get

1. One row per review

App ID, app name, country, star rating, title, body, author, app version, review date, review URL.

2. One app report per app

• Average rating and rating distribution
• Negative-review share
• Rating by country (this is where the surprises are)
• Top complaint categories with counts and representative quotes
• Top praise categories
• Prioritized recommendations
• Optional Chinese labels (summaryLanguage: "zh")

In Apify Console, the dataset opens with two table views: Reviews (one row per review) and App reports (one row per app).

Complaint categories:

Crashes / stability, Login / accounts, Ads, Pricing / subscriptions, Performance / lag, UI / usability, Missing features, Customer support, Notifications, Content quality, Version / update issues.

Real sample: Notion, 200 reviews, 4 countries

Input: appIds: ["1232780281"], countries us, gb, jp, de, 50 reviews each.

Average rating: 2.59 / 5
Negative share: 55.5%

Rating by country
  us  2.52
  gb  2.82
  jp  2.62
  de  2.40

Top complaints:

Category	Negative reviews
Version / update issues	21
Crashes / stability	14
Missing features	14

Example quote:

"Now I cannot manage my page hierarchy any more??? This makes this app unusable on iPad." — 1 star, US

Recommendations returned:

1. Watch for regressions in new releases. Compare complaint themes before and after the update.
2. Fix crashes and stability first, especially in the latest release.
3. Rank feature requests by frequency. Ship the one that affects retention first.

That is the whole point: 200 reviews in, one decision out.

Real sample: Duolingo, 200 reviews, 4 countries

Average rating: 3.65 / 5
Negative share: 28.0%

Rating by country
  us  3.68
  gb  4.10
  jp  3.26
  de  3.56

Same app, same version, 0.84 stars between the UK and Japan. A single-country review feed would never show you that.

Top complaints:

Category	Negative reviews
Pricing / subscriptions	11
Missing features	7
Version / update issues	7

Output example

{
  "_type": "app_summary",
  "appName": "Notion",
  "countries": ["de", "gb", "jp", "us"],
  "reviewCount": 200,
  "averageRating": 2.59,
  "negativeShare": 0.555,
  "countryBreakdown": [
    { "country": "de", "reviewCount": 50, "averageRating": 2.4, "negativeShare": 0.6 },
    { "country": "gb", "reviewCount": 50, "averageRating": 2.82, "negativeShare": 0.5 }
  ],
  "topComplaints": [
    {
      "category": "update",
      "label": "Version / update issues",
      "count": 21,
      "share_of_negative": 0.189,
      "examples": [
        {
          "rating": 1,
          "country": "us",
          "title": "The new version is unusable",
          "body": "Now I cannot manage my page hierarchy any more."
        }
      ]
    }
  ],
  "unclassifiedNegative": 51,
  "unclassifiedShare": 0.459,
  "recommendations": [
    "Watch for regressions in new releases. Compare complaint themes before and after the update."
  ]
}

unclassifiedNegative is published on purpose: short reviews like "Awful" carry no signal, and pretending otherwise would be dishonest. Their quotes are still in the dataset so you can spot a new issue manually.

Input

{
  "appIds": ["1232780281"],
  "countries": ["us", "gb", "jp", "de"],
  "maxReviewsPerCountry": 100,
  "minRating": 1,
  "maxRating": 5,
  "includeSummary": true,
  "summaryLanguage": "en"
}

Field	Type	Description
appIds	array	Numeric App Store IDs or App Store URLs.
countries	array	Two-letter storefront codes. Default: us, gb, jp, de.
maxReviewsPerCountry	integer	1–500. Apple's public RSS caps at ~500 per country.
minRating / maxRating	integer	Filter by star rating.
since	string	Optional YYYY-MM-DD; skip older reviews.
includeSummary	boolean	Add one app report per app.
summaryLanguage	string	en or zh. Default: en.

Pricing

Pay per event:

• $0.50 per 1,000 reviews (review-scraped)
• $0.01 per app report (app-summary)

No monthly subscription. You only pay for what you collect.

A 4-country, 400-review run costs about $0.20.

Why not the $0.10 / 1,000 scraper?

Because you would still have to read the reviews.

	Cheap review scraper	App Review Radar
Output	Raw rows	Rows + report
Countries	Usually one	Aggregated
"Why did rating drop?"	You figure it out	Complaint categories + quotes
"What do I fix first?"	—	Recommendations
Price	$0.10 / 1k	$0.50 / 1k

You are paying $0.40 more per 1,000 reviews to skip an afternoon of reading.

Use cases

• ASO and keyword research
• Competitor review analysis
• Localization QA: compare the same app across countries
• Version regression monitoring
• Product prioritization from real user feedback
• Feed an LLM or MCP agent with clean, structured review data

Limitations

• Apple's public RSS returns the most recent ~500 reviews per country, not full history.
• Reviews are point-in-time; run on a schedule to track changes.
• Category classification is keyword-based and intentionally conservative. Very short reviews are counted as unclassified.
• Public data only. The Actor does not log in, bypass challenges or access private data.
• iOS App Store only. Google Play is not supported yet.
• Not affiliated with Apple.

FAQ

Do I need an Apple developer account? No. This uses Apple's public iTunes lookup and customer-reviews RSS.

Why only ~500 reviews per country? That is Apple's public RSS window. Add more countries or run more often to widen coverage.

Does it work for Google Play? Not yet. This Actor is iOS App Store only.

Can I get Chinese labels? Yes. Set summaryLanguage to zh.

Can I run it on a schedule? Yes. Use Apify Schedules, the API, MCP, Zapier, Make or n8n.