App Store Reviews Scraper & Analyzer – iOS Complaints AI

Scrape the latest public App Store reviews of any iOS app, yours or a competitor's, from up to 15 countries, and get them classified with AI into bugs, crashes, feature requests, pricing/subscription, UX, ads and performance. Each app gets severity, complaint share per version, real quotes and a ranked list of the fixes users ask for most. No App Store Connect access needed.

Need a done-for-you report? Revunda App Store complaint report, €19: a short PDF with severity, version trend and the 10 fixes that matter most (sample).

Find out what your App Store reviews actually complain about. Give this actor one or more App Store URLs (or numeric app IDs). It reads the latest public customer reviews from Apple's official RSS feed across up to 15 storefronts and classifies each review with AI (TypeSafe Jev) into:

• Bugs, Crashes / freezes, Feature requests, Pricing / subscription, UX / design, Ads, Performance
• 18 specific issues (login problems, lost data, trial charges, aggressive paywall, usage limits, redesign backlash, slow loading, battery drain, unresponsive support…)
• a severity score from 0 (no problem) to 3 (app unusable)

Output

• One review item per review: country, rating, version, date, title, text, categories, issues, severity, method (ai or keywords). Reviewer names are not collected.
• One summary item per app: complaint categories with average severity, most affected versions and example quotes; complaint share per version; and topFixes, the 10 specific issues ranked by number of complaining reviews × severity.

Input

Field	Meaning
apps	App Store URLs or IDs, e.g. https://apps.apple.com/us/app/duolingo-language-lessons/id570060128
countries	storefront codes (default: us, gb, ca, au, de, fr, es, it, nl, br, mx, in, jp, se, pl)
maxReviewsPerApp / maxPerCountry	caps (Apple exposes ~500 most recent reviews per country)
onlyNegative	only 1-3 star reviews
useAi / typesafeApiKey	AI on/off; your own key lifts the 300 AI calls/run cap

Empty input runs a small demo on Duolingo.

Example input

{
  "apps": ["https://apps.apple.com/us/app/duolingo-language-lessons/id570060128"],
  "countries": ["us", "gb", "ca", "au"],
  "maxReviewsPerApp": 200,
  "maxPerCountry": 50,
  "onlyNegative": true,
  "useAi": true
}

Example output (shortened)

One review row per review (a real row from a Duolingo test run):

{"type": "review", "appId": "570060128", "appName": "Duolingo: Learn Languages", "reviewId": "14680863495",
 "country": "us", "rating": 5, "version": "7.144.0", "date": "2026-10-18", "title": "Amazing app",
 "text": "It makes me learn Spanish fun", "categories": [], "issues": [], "severity": 0, "method": "ai"}

Plus one summary row per app (also saved in the key-value store as SUMMARY_<appId>) with these fields: appName, developer, currentVersion, allTimeRating, allTimeRatingCount, reviewsAnalysed, averageStars, complaintReviews, countries, categories[] (reviews, avgSeverity, topVersions, example quotes), versions[] (avgStars, complaintSharePct) and topFixes[] (issue, fix, reviews, avgSeverity, score, examples). Export as JSON, CSV or Excel, or pull it through the Apify API.

Pricing

Pay per event: $0.02 per result in the dataset (each review row and each per-app summary row), plus a tiny per-run start fee. Use onlyNegative and maxReviewsPerApp to keep runs small; 100 negative reviews of one app come to about $2. Current prices are on the Pricing tab.

Limits

Apple's public feed only contains the most recent reviews, not the full history. AI can misclassify single reviews, so check the quotes. Not affiliated with Apple.

Need a done-for-you report?

Prefer a finished PDF over a dataset? Revunda uses the same public App Store reviews and sends the report by email:

• App Store complaint report, €19 one-off: what users complain about, how bad it is, which versions it hits and the 10 fixes that matter most. See a real sample report.
• Competitor Pack, €29: your app plus up to 2 competitors, three full reports and a side-by-side summary of their top complaints.
• 20 ready replies to your 1–2★ reviews, €9: ready to paste into App Store Connect or Play Console.
• Weekly Monitor, €29/month: weekly review digest for your app and up to 2 competitors, theme shifts vs last week and spike alerts. Cancel anytime.
• Android app? Google Play complaint report, €9.

Free examples: App Store and Google Play complaint reports for popular apps · app vs app comparisons. Not sure yet? Run the free complaint check or export App Store or Google Play reviews to CSV first.

Pairs with

• google-maps-review-insights: AI analysis of Google Maps reviews with owner reply drafts
• gbp-suspension-risk-audit: Google Business Profile suspension risk score
• restaurant-review-insights: tag restaurant reviews by complaint type

FAQ

How do I analyze App Store reviews of my app or a competitor?

Paste the App Store URL or numeric app ID, pick countries and run; read the top-fixes list first.

Do I need App Store Connect access?

No. Only public reviews are used, so it works for competitors too.

How many reviews does it read?

Apple exposes a limited number of the most recent reviews per country; the actor reads what is publicly available for each selected country.

Can I export App Store reviews to CSV, Excel or JSON?

Yes. Every run writes a dataset you can download as CSV, Excel, JSON or XML, or read through the Apify API. Schedule the actor to get a fresh export every day or week.

Does it scrape Google Play reviews too?

No, this actor reads iOS App Store reviews only. For Android apps, Revunda has a free Google Play review export to CSV and a Google Play complaint report, €9.

What about Shopify App Store reviews?

Not in this actor. Revunda sells a Shopify App Store category gap report for Shopify app developers, built from merchant reviews.

How much does a run cost?

$0.02 per dataset row (reviews plus one summary per app) and a tiny start fee. With onlyNegative on, you pay only for the 1–3★ reviews.

Can I track an app every week?

Yes. Schedule the actor in Apify Console and connect email, Slack or a webhook. For a finished weekly digest instead, see the Revunda Weekly Monitor.