App Store and Google Play keep their reviews in two different places. If you export them twice, you usually get the same reviews again, and you have to figure out what's new yourself.
What you actually want is one file you can put on a schedule, where later runs only keep the new reviews, and a Slack or Discord message when a 1-star complaint shows up.
I built a paid Apify Actor that does this. Below is a real run from Nov 19, 2026, collecting public WhatsApp reviews from both US stores.
What we're setting up
Both stores, one dataset. A schedule that only saves reviews it hasn't already saved. Optional alerts for reviews that match a star limit or a word list. It reads public listings and public written reviews. It doesn't log in, and it doesn't post replies. It isn't affiliated with Apple or Google.
What the output looks like
The dataset uses the same columns for both stores. Each row is a review or an app listing. The image above shows five rows from that run, with long review text cut short.
The first run saved 30 App Store reviews and 30 Google Play reviews, plus a listing row for each store. The App Store reviews ran from Nov 17 to Nov 18, 2026. The Play reviews were all from Nov 18.
Run the first collection
Paste the apps, set a country, and cap how many reviews you want per app, store, and country. Play also splits that cap by language. This is the input I used, except I pasted a real Discord webhook in place of YOUR_WEBHOOK_URL.
{ "apps": [ "https://apps.apple.com/us/app/whatsapp-messenger/id310633997", "com.whatsapp" ], "countries": ["us"], "languages": ["en"], "sort": "newest", "maxReviews": 30, "newReviewsOnly": true, "includeMetadata": true, "alertWebhookUrl": "YOUR_WEBHOOK_URL", "alertMaxStars": 1, "alertMaxPerRun": 3 } languages only applies to Google Play. Apple reviews follow the country. maxReviews is per app, store, and country, and per Play language.
When the run finishes, open the dataset in Apify and export it as CSV, JSON, or Excel.
Try the review-monitoring example on Apify. The link opens the Actor. Paste the JSON above, swap in your own app, and start a small run. It doesn't load these settings for you.
Collect only new reviews
I used the same input for a second run, about half a minute later, which is what a schedule would do.
It still saved 60 new reviews. The first run hit the 30-review cap on each store before it reached the end of the public feed, so the Actor remembered where it stopped. The second run continued from there and saved 30 more App Store reviews and 30 more Play reviews. None of them overlapped with the first run.
On the App Store, those were older reviews from Nov 16 and 17, 2026, that the first run hadn't reached. On Google Play, some were slightly older, and a few were posted after the first run finished. The cap was still 30 per store, so a later run will keep going.
Send matching reviews to Slack or Discord
No extra charge. Paste a Slack incoming webhook or a Discord webhook into alertWebhookUrl, then set a star limit, a word list, or both. If both are set, a review only has to match one of them. Alerts only fire for reviews saved in that run, so a new-only schedule won't ping you twice about the same review. The default cap is 20 per run, and the maximum is 50. If nothing matches, nothing gets sent. It doesn't reply to the review in the store.
I set the star limit to 1 and the cap to 3:
{ "alertWebhookUrl": "YOUR_WEBHOOK_URL", "alertMaxStars": 1, "alertMaxPerRun": 3 } The first run found 15 one-star reviews and sent one Discord message. The cap was 3, so it covered three of them and noted the other 12. I didn't save a copy of that message, and most of its three reviews weren't in English, so here's the same format with three English 1-star reviews from that run, long text cut short:
3 new matching reviews for WhatsApp and 12 more in the dataset ★☆☆☆☆ 1 star · App Store · WhatsApp · US · 2026-11-17 Banned with no reason: I only use WhatsApp to coordinate... ★☆☆☆☆ 1 star · App Store · WhatsApp · US · 2026-11-17 Delay messages: Messages hardly go or com in ★☆☆☆☆ 1 star · Google Play · WhatsApp · US · 2026-11-18 verification code What happens if a run hits its maximum
A scheduled collector that sorts by newest should stop once it reaches the last review it already saved. If it hits maxReviews first, it hasn't caught up. If it moved the marker to the newest review anyway, the next run would think it was done and skip the ones it missed.
Since version 0.2.10, it doesn't move that caught-up marker. It remembers the oldest review this run saved, and the next run continues from there. That's what happened between the two runs above. The status line on both was: "Reached maxReviews before catching up. The next run will continue from where this one stopped."
When the official tools are enough
For your own apps, start with the official dashboards.
Google Play Console lets you download monthly CSV review reports. You pick an app, a year, and a month, so one download isn't necessarily the entire history. There's also a reviews API, but it only returns reviews created or edited in the last week. If a sync job dies and nobody notices for eight days, those reviews are gone from the API unless someone edits them later.
App Store Connect is the official dashboard for your own iOS app, and Apple has a Customer Reviews API you can page through. I haven't found a one-click CSV export there.
For public reviews of apps you don't own, Apple has a public RSS/JSON feed. No API key needed. Each page has 50 reviews. I asked for page 11 of WhatsApp on the US storefront on Nov 19, 2026, and got HTTP 400 with CustomerReviews RSS page depth is limited to 10. So you get about 500 written reviews per country, per sort, and that's it. Google doesn't offer a public reviews API for other people's apps.
Limits and costs
Apple's public window is that roughly 500-review cap per country. A long gap on a busy app can miss reviews that have already fallen out of it. Countries are storefronts, not languages, and Play reviews also follow the language list. Star and word filters drop non-matching reviews before they're saved, and you don't pay for those.
List prices are $0.001 per run, $0.00012 per review ($0.12 per 1,000), and $0.002 per listing row.
The first run's charge, computed from those list prices and the events it saved, is $0.0122: $0.001 to start, 60 × $0.00012 for reviews, and 2 × $0.002 for the two listings. That's computed from list prices, not an invoice. The second run saved another 60 reviews and both listings again, because I left "skip an unchanged listing" off, so the same list-price total applies again.
Try the example
If you want this workflow, open the Actor, paste the JSON above, replace WhatsApp with your app, and run a small collection so you can inspect the output.
Try the review-monitoring example on Apify
If something's broken or missing, tell me.




![Physical AI Hardware Software Co Design — Ch.00: Introduction [EN]](https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Ftzwmyyh3qv11u91ly0vs.png)

