Booking.com Reviews Scraper & API: All Reviews, from $0.50/1K avatar

Booking.com Reviews Scraper & API: All Reviews, from $0.50/1K

Pricing

from $0.50 / 1,000 reviews

Go to Apify Store
Booking.com Reviews Scraper & API: All Reviews, from $0.50/1K

Booking.com Reviews Scraper & API: All Reviews, from $0.50/1K

Scrape Booking.com guest reviews of any property: score, title, liked and disliked text, room, nights, stay dates, traveller type, country and property reply, plus address, coordinates and score. No page limit, filters for language and traveller type, new-reviews-only monitoring and webhook.

Pricing

from $0.50 / 1,000 reviews

Rating

0.0

(0)

Developer

Alexis Morain

Alexis Morain

Maintained by Community

Actor stats

0

Bookmarked

1

Total users

0

Monthly active users

4 days ago

Last modified

Features

Limited permissions

Share

Booking.com Reviews Scraper

This Actor exports the guest reviews of any property on Booking.com. Give it property URLs or hotel ids, and it returns one row per review: the guest score out of 10, the title, what the guest liked and what the guest disliked as separate fields, the room, the number of nights, the check-in and check-out dates, the traveller type, the guest's country and the property's reply. A summary row per property adds its address, coordinates, star rating and overall score.

It works as a Booking.com reviews API. Booking protects its property pages with a bot check, so most scrapers open every property in a full browser. This Actor reads the reviews from Booking's review endpoint over plain HTTP, 25 reviews per request. It uses a headless browser for a few seconds only to turn a property URL into its numeric hotel id (and caches the id), and it does not use one at all if you paste the id. Pricing is pay per event: $1.00 per 1,000 reviews on the Apify Free plan, $0.50 per 1,000 on Gold, plus $0.005 when a run starts.

What makes it different: no page limit (22,271 reviews of one hotel in one run), liked and disliked text in separate columns, filters for language and traveller type applied by Booking itself, and a New reviews only mode with a webhook for scheduled monitoring.


What does Booking.com Reviews Scraper do?

For each property you give it, the Actor:

  1. Finds the numeric hotel id. If you gave an id, this step is skipped. If you gave a URL, the id comes from the cache of earlier runs, or from a headless browser that passes Booking's bot check.
  2. Reads the property's name, address, coordinates, star rating, overall score and review count.
  3. Reads the reviews newest first, 25 per request, four requests at a time, until the last review Booking shows.
  4. Pushes the reviews in batches of 100, and adds one summary row for the property unless you turn it off.

Why scrape Booking.com reviews?

  • Reputation monitoring for hotels and apartments. Schedule a daily run in New reviews only mode, with a webhook to Slack, Make, Zapier or n8n.
  • Competitor research. Read what guests dislike at the properties around yours. The negativeText column holds only the complaint, so you can count themes without parsing a full review.
  • Revenue and product decisions. Group scores by room type, traveller type, nights stayed or guest country.
  • Sentiment analysis and LLM pipelines. Score, language code and the positive and negative text per review.
  • Market studies. The summary row carries address, latitude, longitude, stars and score, so a list of properties becomes a map.

Apify adds scheduling, an API for every run, integrations with Google Sheets, Slack, Zapier, Make and n8n, and exports to JSON, CSV and Excel.

How to scrape Booking.com reviews

From the Apify console

  1. Click Try for free and open the Actor.
  2. Open a property on Booking.com and copy its URL, for example https://www.booking.com/hotel/fr/le-six.html. Language suffixes such as .en-gb.html and query strings are fine.
  3. Paste it in Page URLs, one per line. You can also paste a numeric hotel id such as 58824.
  4. Set Max reviews per page URL. The default is 100, and 0 means every review.
  5. Optionally filter by language, traveller type or date.
  6. Click Start, then export the dataset as JSON, CSV or Excel.

How to find the hotel id

Run the Actor once with the URL. The placeId of the summary row is the numeric hotel id. Use it as input next time and no browser starts. (The Actor also caches the URL to id lookup in a named key-value store of your account, booking-hotel-ids, so repeat runs on the same URL skip the browser as well.)

Which fields should I fill?

I want to...FillNotes
Try it oncePage URLsThe prefilled URL with 20 reviews costs about three cents on the Free plan.
Get every review Booking showsMax reviews per page URL = 0One charged result per review, so this is also your cost cap per URL.
Skip the browser stepHotel id instead of URLFastest start.
Get recent reviewsOnly reviews from the last N days1 to 3650. Reads newest first and stops when a whole page is older than the date.
Get reviews since a fixed dateOnly reviews published afterYYYY-MM-DD. If both are set, the later date wins.
Keep some languagesReview languagesCodes such as en, fr, de, es, it, nl, pt. Empty means all.
Keep some traveller typesTraveller typesFAMILY, COUPLES, SOLO, BUSINESS, FRIENDS. Filtered by Booking itself.
Monitor a propertyReview mode = New reviews onlyAdd a Monitoring key for several independent watches on one property.
Be told when a run endsWebhook URLReceives a JSON summary.
Get a reviews-only CSVAdd one summary row per URL = offOtherwise the dataset also holds one place row per property.

Input examples

One property, the 100 newest reviews:

{
"startUrls": ["https://www.booking.com/hotel/fr/le-six.html"],
"maxReviewsPerPlace": 100
}

Everything Booking shows for a large hotel, by hotel id:

{
"startUrls": ["58824"],
"maxReviewsPerPlace": 0
}

Recent English and French reviews of two properties:

{
"startUrls": [
"https://www.booking.com/hotel/gb/park-plaza-westminster-bridge.html",
"58824"
],
"maxReviewsPerPlace": 0,
"languages": ["en", "fr"],
"lastDays": 90
}

Solo and business travellers only:

{
"startUrls": ["https://www.booking.com/hotel/fr/le-six.html"],
"maxReviewsPerPlace": 0,
"travelerTypes": ["SOLO", "BUSINESS"]
}

Daily monitoring, with a webhook:

{
"startUrls": ["58824"],
"maxReviewsPerPlace": 0,
"reviewMode": "new-only",
"webhookUrl": "https://hooks.example.com/booking"
}

Details that need a word

  • The review is split in two. Booking has a "liked" box and a "disliked" box. They come out as positiveText and negativeText. The text column is always null on Booking rows, and title is null when the guest wrote none.
  • English is en. Booking marks English reviews with an internal code (xu). The Actor reports them as en, and the en filter gives the same count as Booking's own.
  • Several traveller types are read one after the other. Booking accepts one type per request, so the Actor makes one read per type. A review has one type, so nothing is returned twice.
  • A URL that redirects is skipped. A property that has been closed redirects to its city page. The Actor reports not a live Booking.com property for it and moves on. In our test run one such URL failed alone and the other two were read (22,847 reviews). The run still counts as succeeded when at least one URL works.
  • If the bot check refuses the browser step, the Actor says so and suggests the numeric hotel id, which always works. It retries the lookup once on another IP before giving up.
  • Guest names. reviewerName is the name Booking shows, usually a first name. It is null when the guest reviewed anonymously.

What data can it extract? Every output field

Each review is one dataset row with "type": "review".

FieldTypeDescription
typestringreview for a review, place for the summary row
sourcestringAlways booking
reviewIdstringBooking review id, stable across runs
ratingnumberGuest score, out of 10
ratingScaleintegerAlways 10
titlestringReview title, when the guest wrote one
positiveTextstringWhat the guest liked
negativeTextstringWhat the guest disliked
languagestringLanguage code of the review
publishedAtstringReview date and time, ISO 8601 UTC
stayDatestringMonth of the stay, YYYY-MM
checkinDatestringCheck-in date, YYYY-MM-DD
checkoutDatestringCheck-out date, YYYY-MM-DD
nightsintegerNights stayed
travelerTypestringFAMILY, COUPLES, FRIENDS, SOLO or BUSINESS
roomTypestringRoom the guest booked
helpfulVotesintegerHelpful votes on the review
photosarrayURLs of the guest's photos (largest size Booking serves), empty when none
reviewerNamestringFirst name or username, null when anonymous
reviewerLocationstringGuest country name
reviewerCountryCodestringGuest country code, upper case
ownerResponsestringThe property's reply
placeIdstringNumeric Booking hotel id
placeNamestringProperty name
placeTypestringAlways hotel on Booking rows
placeUrlstringProperty page on Booking.com
providerstringAlways Booking.com
scrapedAtstringWhen the row was collected, ISO 8601 UTC

text, reviewUrl, originalLanguage, subRatings, reviewerProfileUrl, reviewerContributions, ownerResponseDate and ownerResponderRole are null on Booking rows. They exist to keep the column names of the Tripadvisor and Agoda Actors. isMachineTranslated is always false.

The summary row

One row per property, with "type": "place":

FieldDescription
placeId, placeName, placeType, placeUrlThe property
address, city, countryCodeWhere it is
latitude, longitudeCoordinates
starsStar rating
rating, ratingScaleOverall Booking score out of 10
reviewsCountNumber of reviews Booking reports for the property
reviewsMatchingReviews the review endpoint reports for the run's filters
reviewsDeliveredReviews this run delivered
inputThe URL or id you gave
scrapedAtCollection time

Comparing reviewsCount with reviewsDelivered tells you how much Booking's review list holds compared with the number on the property page.

Example output

A review row (shortened; the place fields are left out):

{
"type": "review",
"source": "booking",
"reviewId": "114ec939d56e5f91",
"rating": 8.0,
"ratingScale": 10,
"title": null,
"positiveText": "מיקום טוב, קרוב ל 3 תחנות מטרו ורחוב ראשי עם מסעדות רבות. מלון חדש מודרני",
"negativeText": "א. בוקר בסיסית\nיום אחד מתוך 4 לילות לא ניקו את החדר",
"language": "he",
"publishedAt": "2026-10-21T20:39:17Z",
"stayDate": "2026-10",
"checkinDate": "2026-10-17",
"checkoutDate": "2026-10-21",
"nights": 4,
"travelerType": "SOLO",
"roomType": "Classic Double Room",
"helpfulVotes": 0,
"photos": [],
"reviewerName": "Aviram",
"reviewerLocation": "Israel",
"reviewerCountryCode": "IL",
"ownerResponse": null,
"provider": "Booking.com"
}

The reviews keep the language the guest wrote in: this one is Hebrew, and language says he. Booking does not translate them in this output.

A summary row (address, coordinates and stars are also filled, here left out):

{
"type": "place",
"source": "booking",
"placeId": "58824",
"placeName": "Hotel Le Six",
"placeType": "hotel",
"ratingScale": 10,
"reviewsMatching": 576,
"reviewsDelivered": 576,
"input": "https://www.booking.com/hotel/fr/le-six.html"
}

How many Booking.com reviews can it get? Measurements

Measured on the Apify platform on 21 October 2026, with the datacenter proxy:

PropertyReviews on BookingDelivered
Park Plaza London Westminster Bridge22,27122,271
Hotel Le Six, Paris576576

Both came from one run with three URLs (the third, a property that no longer exists, failed alone). The run delivered 22,847 reviews in 188 seconds, and that time includes the browser step that turned the three URLs into hotel ids.

In the recon before the Actor was built, the Excalibur hotel gave 30,655 of 30,655 reviews in 1,229 requests and 116 seconds, four requests in parallel. No page limit showed up.

Booking only shows about three years. Booking.com publishes reviews from the last 36 months (the state of the source in our recon of 21 October 2026). The Actor returns what Booking shows, so that is the history you get. If reviewsCount in the summary row is higher than the reviews delivered, that is where the gap comes from.

Monitor new Booking.com reviews: New reviews only mode

Set Review mode to New reviews only and schedule the Actor.

  • The Actor keeps a registry per property in a named key-value store of your account, booking-reviews-registry. It holds the ids of the reviews already delivered.
  • Set Max reviews per page URL to 0 on the first run. A run stopped by the review limit, a spend limit, a timeout or unreadable pages is not recorded as a complete read, so the next run reads the full history again. Reviews already delivered are never returned or charged twice.
  • After a complete read, later runs read only the reviews newer than that read minus 30 days. The margin covers reviews that appear after newer ones (moderation).
  • Skipped reviews are not charged. The status message says how many were skipped.
  • The registry is tied to the filters (languages, traveller types). Another set of filters is another watch. A Monitoring key gives the watch a name of your choice; changing the filters under the same key makes the next run read the full history once.
  • If you set a date window (last N days or published after), the run is not recorded as a complete read.

Webhook at the end of the run

Fill Webhook URL and the Actor sends one JSON POST when it has finished reading (example values):

{
"actor": "booking-reviews-scraper",
"runId": "abc123",
"datasetId": "def456",
"mode": "new-only",
"status": "succeeded",
"places": 1,
"reviews": 2,
"skippedAlreadySeen": 575,
"targets": 1,
"failedTargets": 0,
"chargeLimitReached": false
}

status is succeeded, partial (some URLs failed) or failed (every URL failed). The URL is stored as a secret and is never written to the log. The call has a 15 second timeout, a 307 or 308 redirect is followed once, and a failed call never changes the run's result. The webhook is not sent when the input is invalid.

Export Booking.com reviews to CSV, Excel or JSON

Open the run, go to Storage or the Output tab, and choose JSON, CSV, Excel, HTML or XML. From the API:

https://api.apify.com/v2/datasets/<DATASET_ID>/items?format=csv&token=<YOUR_API_TOKEN>

For a file with reviews only, turn off Add one summary row per URL, or filter the type column after the export. The photos field is a list: it stays intact in JSON.

Using the Booking.com reviews API from your code

Python

from apify_client import ApifyClient
client = ApifyClient("<YOUR_API_TOKEN>")
run = client.actor("alexmorain/booking-reviews-scraper").call(run_input={
"startUrls": ["https://www.booking.com/hotel/fr/le-six.html"],
"maxReviewsPerPlace": 0,
"languages": ["en"],
})
for row in client.dataset(run["defaultDatasetId"]).iterate_items():
if row["type"] == "review" and row["negativeText"]:
print(row["rating"], row["roomType"], row["negativeText"])

JavaScript

import { ApifyClient } from 'apify-client';
const client = new ApifyClient({ token: '<YOUR_API_TOKEN>' });
const run = await client.actor('alexmorain/booking-reviews-scraper').call({
startUrls: ['58824'],
maxReviewsPerPlace: 0,
lastDays: 30,
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();
const reviews = items.filter((row) => row.type === 'review');
console.log(reviews.length, 'reviews in the last 30 days');

cURL

curl -X POST "https://api.apify.com/v2/acts/alexmorain~booking-reviews-scraper/run-sync-get-dataset-items?token=<YOUR_API_TOKEN>" \
-H "Content-Type: application/json" \
-d '{"startUrls": ["58824"], "maxReviewsPerPlace": 50}'

This call waits for the run and returns the items. For large properties, start the run with POST /v2/acts/alexmorain~booking-reviews-scraper/runs and read the dataset when it ends.

What people use it for

Hotel reputation monitoring

List your properties by hotel id, schedule a daily New reviews only run and send the webhook to a chat channel. A run with a few new reviews costs about one cent on the Free plan.

Finding what guests dislike near you

Run the Actor on the properties in your area and read negativeText, with roomType and travelerType next to it. Count the recurring words: breakfast, noise, cleaning.

Score analysis by guest origin and room

reviewerCountryCode, roomType, nights and travelerType are separate columns, so a pivot table gives the average score by country or room.

Property datasets

The summary row gives name, address, coordinates, stars and score for each property you list, next to its reviews.

Booking.com Reviews Scraper pricing

You pay per event. There is no monthly rental.

EventFree planBronzeSilverGold and above
Review (per 1,000)$1.00$0.90$0.80$0.50
Summary row (per 1,000)$2.00$1.80$1.60$1.00
Actor start (per run, per GB of memory, minimum one)$0.005$0.005$0.005$0.005

The Actor runs with 1,024 MB (the browser step needs it), which is one GB, so the start fee is $0.005 per run. The tier that applies is the one of your Apify plan.

JobFree planGold plan
100 reviews, one property$0.107$0.056
1,000 reviews, one property$1.007$0.506
All 576 reviews of Hotel Le Six$0.583$0.294
All 22,271 reviews of Park Plaza Westminster Bridge$22.28$11.14
Daily monitoring run with 3 new reviews$0.010$0.0075

Each worked example is the reviews plus one summary row ($0.002 Free, $0.001 Gold) plus the start fee. The $5 of monthly credit in the Apify Free plan covers about 5,000 reviews.

Cost control. Max reviews per page URL caps each URL. You can also set a maximum cost for the run in the Apify run options. The Actor pushes reviews through the platform's charging check, so it stops at the limit instead of passing it; we checked this on the platform before publishing. Reviews already delivered in New reviews only mode are not charged again.

Limits and what this Actor does not do

  • About 36 months of history. Booking.com shows the last 36 months of reviews, and the Actor returns what Booking shows.
  • Property URLs must contain /hotel/<country>/. City pages and search results are not read.
  • A closed property cannot be read. Its URL redirects to a city page.
  • The URL to id step needs a browser. It adds a few seconds on the first run for a URL, and it can be refused by Booking's bot check. A numeric hotel id avoids both.
  • No translation. Reviews keep the language they were written in.
  • The title is empty on some reviews. The example row above has none.
  • No per-review sub-scores (cleanliness, staff and so on). The Actor does not return them.
  • No reply date or replier role. Only the text of the property's reply.
  • Proxy. Booking answers a challenge to direct Apify IPs, so the Actor needs the Apify datacenter proxy. Keep the Proxy field at its default.

FAQ

Is there a Booking.com reviews API?

This Actor does not use an official Booking.com API. It reads the review data that the public property pages load and returns it as JSON through the Apify API. You call it with an Apify token.

How do I scrape Booking.com reviews?

Paste the property URL or hotel id in the Actor, set the maximum number of reviews, press Start, and download the dataset as CSV, Excel or JSON.

How do I export Booking.com reviews to CSV?

Run the Actor, open the dataset and choose CSV. Turn off the summary row if you want only reviews in the file.

How many reviews can I scrape from one property?

All the reviews Booking shows. We read 22,271 of 22,271 on one hotel. Set Max reviews per page URL to 0.

Why do I get fewer reviews than the number on the property page?

Booking shows the last 36 months in its review list, and the number on the property page can be higher. Language and traveller filters also reduce the count. The summary row gives both numbers.

Can I scrape the negative reviews only?

Booking has no star filter in this Actor. Filter the dataset on rating after the run, or read the negativeText column, which holds only what the guest disliked.

Can I get the reviews in English only?

Yes. Fill Review languages with ["en"]. Booking applies the filter.

What is a Booking.com hotel id and where do I find it?

It is the number Booking uses internally for a property. Run the Actor once with the URL, and read placeId in the summary row.

Does it need a Booking account or a browser?

No account. A browser is used only for a few seconds, to find the hotel id of a URL. Paste an id and no browser starts.

What if a URL fails?

The log says why and the Actor moves to the next URL. A closed property redirects to a city page and is reported as not a live property. If the bot check refuses the id lookup, use the numeric hotel id.

How do I monitor new Booking.com reviews?

Use New reviews only with a daily or weekly schedule and add a webhook. See the monitoring section.

The Actor reads pages that anyone can open without logging in. The output contains data that guests made public: a first name or username, a country, the review text. Personal data is protected by the GDPR in the European Union and by similar laws elsewhere. Use the data for a legitimate purpose, check Booking.com's terms and take legal advice for your own use. This is not legal advice.

Integrations, exports and support

Connect the dataset to Google Sheets, Slack, Zapier, Make or n8n with Apify's integrations, or use the Apify API and client libraries. Runs can be scheduled from the console.

Something not working, or a field you need? Open an issue in the Issues tab. The same developer publishes Tripadvisor Reviews Scraper, Agoda Reviews Scraper and Airbnb Reviews Scraper, with the same column names.

Guide and use cases

Use cases for outbound and automation, the output fields, pricing and common questions are on the Booking.com Reviews Scraper page of my site. The Reviews family page puts it side by side with the other scrapers of the same family.