App Reviews Scraper — Google Play and App Store

Export written app reviews from both stores in one dataset, with app metadata on every review. Use Play packages, App Store ids, or store URLs in the same input list. No login required.

Quick start

{
  "apps": ["com.whatsapp", "310633997"],
  "maxReviewsPerApp": 100,
  "sort": "newest",
  "country": "us",
  "language": "en"
}

Each review includes the app title, developer, average rating, rating count and version. Google Play also includes installs, helpful votes and developer replies when available. Export JSON, CSV or Excel from the dataset. Compare written feedback across Android and iPhone, track new complaints, or feed review text into your own analysis workflow.

Input

Field	Default	Behavior
apps	Required	Play package (com.whatsapp), Play details URL, App Store id (310633997 or id310633997), or App Store URL. Up to 100 inputs, deduplicated by store + id.
maxReviewsPerApp	100	1–10,000 on Google Play. App Store caps at 500 per app and country.
sort	newest	newest or most_relevant (Play helpfulness, App Store mosthelpful).
country	us	Two-letter storefront country. One country per run; run again for another country.
language	en	Google Play only. App Store review language is not selected by this field.

Default Store example requests only 10 reviews from each of two apps. App metadata is fetched once per app. Two apps run concurrently; each app's pages are fetched sequentially.

Output

One flat row per written review. All keys are always present; absent strings are "", absent numbers are 0. No nulls.

Fields	Meaning
inputIndex, input	Original input position and value; duplicate apps are removed.
store, appId	google_play or app_store, and normalized package/id. Invalid inputs have empty store/id.
appTitle, developer, appRating, appRatingsCount	App metadata from the selected country.
appInstalls, appVersion	Install range on Play; installs is empty on App Store. Current app version.
reviewId, userName, rating	Review identity, public display name and 1–5 score.
title, text, date	Written review and UTC ISO date. Title is empty on Play.
reviewedVersion, helpfulCount	Reviewed app version; Play thumbs-up or App Store vote count.
replyText, replyDate	Play developer reply when supplied. Empty on App Store's public feed.
url	Play review URL; App Store app reviews page (the feed does not expose a stable review permalink).
country, language	Storefront country; language is empty for App Store because the RSS feed does not report it.
status, subStatus, error, scrapedAt	Result status, machine-readable detail, error and UTC collection timestamp.

A failed app produces a free failure row. If a later page fails after reviews were saved, those review rows remain and a free terminal failure row explains the stop.

status / subStatus	Charged review event?
ok / review	Yes
no_reviews / no_reviews	No; app exists but returned no written reviews
failed / invalid_input	No
failed / not_found	No; app missing in the selected country
failed / http_CODE, rate_limited, network_error	No
failed / free_work_limit	No; internal guard stopped unproductive work
failed / cost_limit_reached	No; your event spending cap stopped the run

Pricing

One review event per saved written review; app metadata included. FREE $0.0002, BRONZE $0.00015, SILVER $0.00012, GOLD and above $0.0001 per review ($0.20 / $0.15 / $0.12 / $0.10 per 1,000). Standard Actor start event: $0.00005 per GB of memory, minimum one event. Failed and empty rows have no review charge; the start event still applies. No extra dataset-item fee and no separate app lookup fee.

Rows are saved before review events are charged. The actor honors your maximum charge setting and bounds free work. A successful 4,000-review Apify spike used 256 MB with no proxy; this is a feasibility measurement, not a guaranteed throughput promise.

Honest limits

• App Store public RSS exposes at most 10 pages / 500 reviews per app and country, ordered newest or most helpful. Asking for more logs the cap and returns at most 500. For other storefront feedback, run separately with another country; it does not bypass the per-country limit.
• Both feeds contain written reviews, not the full population of star ratings. Returned review count can differ from the app's overall rating/review counters.
• Google Play depends on public web endpoints through google-play-scraper 10.1.4, pinned exactly. Store changes can break parsing or pagination.
• Feed availability, country coverage, pagination and helpfulness order are controlled by the stores. There is no guarantee of all historical reviews, a particular count, or identical results across runs.
• App Store developer replies and install counts are not exposed by these endpoints. We do not extract private bearer tokens or use amp-api.
• Only public endpoints are used. No login, residential proxy or user-supplied cookie. Play 429 responses get one direct retry after backoff, then at most one Apify datacenter proxy attempt if available; persistent failures return free rows.
• Requests have hard deadlines; an internal cost guard can stop repeated failures with free_work_limit. No sentiment analysis or translation is performed.

API examples

Use your own Apify token via an environment variable. Replace the placeholder locally.

curl -X POST \
  "https://api.apify.com/v2/acts/sumitr_mardy~app-reviews-scraper/run-sync-get-dataset-items" \
  -H "Authorization: Bearer $APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"apps":["com.whatsapp","310633997"],"maxReviewsPerApp":10,"country":"us"}'

import os
from apify_client import ApifyClient

client = ApifyClient(os.environ["APIFY_TOKEN"])
run = client.actor("sumitr_mardy/app-reviews-scraper").call(
    run_input={"apps": ["com.whatsapp", "310633997"], "maxReviewsPerApp": 10}
)
for review in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(review["store"], review["rating"], review["text"])

Use it from AI agents (MCP)

Hosted clients supporting OAuth can connect to the Apify MCP server and sign in with Apify.

{
  "mcpServers": {
    "app-reviews": {
      "url": "https://mcp.apify.com/?tools=sumitr_mardy/app-reviews-scraper"
    }
  }
}

For clients using Bearer authentication, supply your token through local secret configuration:

{
  "mcpServers": {
    "app-reviews": {
      "url": "https://mcp.apify.com/?tools=sumitr_mardy/app-reviews-scraper",
      "headers": {"Authorization": "Bearer <YOUR_APIFY_TOKEN>"}
    }
  }
}

For stdio clients with Node.js:

{
  "mcpServers": {
    "app-reviews": {
      "command": "npx",
      "args": ["-y", "@apify/actors-mcp-server", "--actors", "sumitr_mardy/app-reviews-scraper"],
      "env": {"APIFY_TOKEN": "<YOUR_APIFY_TOKEN>"}
    }
  }
}

Agent tips: start with 10 reviews per app, inspect status before analyzing text, use the same country when comparing stores, and report the App Store cap. Never treat failure rows as customer reviews. For recurring collection, use your own Apify schedule and deduplicate by store + appId + country + reviewId.

Support: use the Actor issues page.