Apple App Store Localization Scraper & Gap Finder

Find Apple App Store localization gaps with real public data. Search a niche or inspect known App IDs, then get one explainable analysis per app with missing languages, rating volume, review evidence, release freshness, confidence, and an opportunity score.

Use this Actor for App Store localization research, ASO market discovery, competitor audits, and country-specific review analysis—without an Apple Developer API key.

Use the evidence to prioritize iOS apps or niches for deeper localization, competitor, ASO, or product research.

Run a market scan on Apify

Watch the 5-minute tutorial

See the complete habit tracker workflow: configure a US market search, check French, German, and Spanish coverage, request up to 25 apps, and interpret a real result.

Verified tutorial run: habit tracker, US storefront, 25 requested, 23 unique apps returned. One real result scored Finch at 74.63 with French, German, and Spanish missing from its declared language list.

What you get

Question	Evidence returned
Which established apps may be under-localized?	Missing target languages plus rating volume and quality
Why did an app receive its score?	Five score components, reasons, caveats, and confidence
Are users asking for another language?	Matching country-specific reviews and raw review evidence
Is the result complete?	Explicit complete, partial, empty, budget_limited, or failed status
Can I reuse the data?	Stable app-level rows exportable as JSON, CSV, or Excel

Ratings and reviews are kept separate: rating volume measures App Store traction, while review evidence measures matched localization pain. One dataset row always represents one unique app—not an unpredictable bundle of paid review rows.

Choose a workflow

Workflow	Use it when	Immediate result
Search a market	You have a niche or keyword	Ranked apps, missing languages, score, and confidence
Analyze reviews	You have App IDs and want country-level pain evidence	Matching reviews, raw evidence, and complete/partial status
Check known apps	You already know the competitors	Side-by-side metadata and localization-gap analysis

Quick start: find App Store localization gaps

1. Open the Actor and choose Search a market.
2. Enter a niche such as habit tracker.
3. Keep the App Store country as us.
4. Add target languages fr, de, and es, then keep 25 apps.
5. Leave review analysis off for the fastest first scan and click Start.
6. Open the dataset and compare opportunityScore, missingLanguages, ratingsCount, confidenceLevel, and caveats.

Equivalent input:

{
  "mode": "search",
  "searchTerm": "habit tracker",
  "country": "us",
  "targetLanguages": ["fr", "de", "es"],
  "maxResults": 25,
  "includeReviews": false
}

The empty input {} is also useful: it searches meditation in the US App Store, analyzes up to 25 apps, and checks whether French is missing.

Equivalent default input:

{
  "mode": "search",
  "searchTerm": "meditation",
  "country": "us",
  "targetLanguages": ["fr"],
  "maxResults": 25,
  "includeReviews": false
}

Example output

This is a shortened record from the real tutorial run recorded on 2026-07-30. Apple returned 23 unique apps when the requested maximum was 25; results can change as Apple updates its public data.

{
  "appId": "1528595748",
  "name": "Finch: Self-Care Pet",
  "dataStatus": "complete",
  "opportunityScore": 74.63,
  "opportunityVerdict": "promising",
  "confidenceLevel": "medium",
  "evidence": {
    "rating": 4.95,
    "ratingsCount": 733046,
    "targetLanguages": ["fr", "de", "es"],
    "missingLanguages": ["fr", "de", "es"],
    "reviewsFetched": 0,
    "matchingReviewCount": 0,
    "reviewEvidenceStatus": "not_requested"
  },
  "scoreComponents": {
    "tractionVolume": 25,
    "ratingQuality": 14.63,
    "localizationGap": 25,
    "reviewEvidence": 0,
    "freshness": 10
  },
  "caveats": [
    "The score does not estimate downloads, revenue, or product-market fit.",
    "Review evidence was not requested."
  ]
}

Pricing

Billing unit	Current configured event price
Actor start	$0.00005
Opportunity analysis	$5 / 1,000 analyzed apps

That is $0.005 per analyzed app. A search that produces 25 unique app rows costs up to $0.125 in analysis events, plus the Actor start event.

Check the Actor price card before starting a run; it is the source of truth if pricing changes.

Reviews are supporting evidence stored separately in the default key-value store. They do not create extra paid dataset rows. One paid default-dataset item equals one unique app opportunity analysis.

Input examples

The Actor input form marks each field as required, recommended, or optional. The three examples below are ready for API calls, Tasks, schedules, or integrations.

Search and rank a niche with review evidence

{
  "mode": "search",
  "searchTerm": "habit tracker",
  "country": "us",
  "targetLanguages": ["fr", "de"],
  "maxResults": 50,
  "includeReviews": true,
  "reviewCountry": "fr",
  "reviewPages": 2,
  "maxReviewRequests": 30,
  "maxConcurrency": 5,
  "filterKeywords": ["traduction", "français", "anglais", "langue"]
}

Analyze App Store reviews by country

{
  "mode": "reviews",
  "appIds": ["961633456"],
  "country": "us",
  "reviewCountry": "fr",
  "targetLanguages": ["fr"],
  "reviewPages": 1,
  "maxReviewRequests": 10,
  "filterKeywords": ["traduction", "anglais", "français"]
}

Audit known iOS apps by App ID

{
  "mode": "lookup",
  "appIds": ["337472899", "324684580"],
  "country": "us",
  "targetLanguages": ["fr", "de", "es"]
}

How the opportunity score works

Every app receives a deterministic score from 0 to 100:

Component	Maximum
Rating volume	25
Rating quality	15
Missing target languages	25
Direct localization-pain evidence in reviews	25
Release freshness	10

The output includes every component, underlying counts, reasons, caveats, evidence coverage, and confidence level.

The opportunity score is not a revenue or product-market-fit prediction. It prioritizes apps for further research using evidence available from Apple's public endpoints. It does not estimate downloads, revenue, local competition, implementation difficulty, or operating cost.

Output and run report

The default dataset has a stable grain: one unique app per row.

Field	Meaning
appId	Unique Apple App Store ID
dataStatus	complete, partial, empty, or budget_limited
opportunityScore	Deterministic score from 0 to 100
opportunityVerdict	promising, investigate, or weak
confidence	Evidence-coverage score and level
scoreComponents	Contribution of every scoring component
evidence	Ratings, missing languages, fetched reviews, and matches
reasons	Human-readable explanation
caveats	Missing evidence and interpretation limits
reviewEvidenceKey	Key-value-store record containing raw reviews
billing	Explicit unit and billable-unit count

When reviews are requested, the Actor writes the full array to a record such as REVIEWS_961633456_FR in the default key-value store and references it from the paid analysis row.

Every run also writes a REPORT record with:

• final complete, partial, empty, budget_limited, or failed status;
• apps discovered, deduplicated, and analyzed;
• review requests attempted, succeeded, failed, and skipped by the cap;
• reviews fetched and duplicates removed;
• paid dataset items and charge-limit status;
• bounded error details.

An Apple error therefore cannot silently look like a complete empty run.

Review request and cost controls

Review extraction is bounded by three controls:

• reviewPages: maximum 3 review pages per app;
• maxReviewRequests: hard global cap across the whole run;
• maxConcurrency: maximum active review requests.

The default cap is 25 review requests and the absolute schema maximum is 100. There is exactly one HTTP attempt per review page—there are no hidden retries behind the request counter. Empty pages stop early. When the user's Apify charge limit is reached, the Actor stops producing paid analyses. If billing capacity cannot be checked, optional review work fails safe and the run is explicitly marked partial.

The runtime is fixed at 256 MB. Metadata requests have a 15-second timeout and make up to three attempts total; review pages have a 12-second timeout and one attempt. The conservative one-app upper bound is 81 seconds. This is a guardrail model, not a substitute for checking actual paid-run margins in Apify Analytics.

For the current 256 MB runtime, the conservative one-app model is 45 seconds of metadata attempts plus up to three 12-second review-page attempts. At the current $0.005 analysis price, an 80% creator share is $0.004 per analyzed app. After the five-second start allowance, 76 billable compute seconds cost about $0.00106 at $0.20 per compute unit, before small storage and transfer costs. This is a guardrail model, not a promise of margin; verify actual paid-run economics in Apify Analytics.

Automate App Store research

Use Apify schedules and integrations to create:

• a recurring App Store localization-gap watchlist;
• competitor release and language monitoring;
• country-specific localization-pain evidence;
• exported dataset inputs for Google Sheets, Airtable, Make, n8n, or an LLM workflow.

The stable app-level dataset makes snapshots safe to compare without mixing app rows and review rows.

Data sources and limitations

• Uses Apple's public iTunes Search, Lookup, and customer-review feeds.
• No Apple Developer API key is required.
• Apple can return fewer apps than requested.
• Review availability and freshness depend on the public storefront feed.
• Declared language support does not prove that every screen or support flow is fully localized.
• The Actor does not claim downloads or revenue because Apple does not expose those metrics through these public endpoints.
• Legal, jurisdiction, licensing, permission, compliance, and Terms of Service questions are excluded from the opportunity score and handled separately by the operator when useful.

Support

Open an issue from the Actor page with the run URL and the REPORT record. Do not include private credentials.