Booking Reviews Scraper

Collect guest reviews from Booking.com hotel pages into a clean dataset. Add one or more hotel URLs, choose how many reviews you want, and run the Actor.

What this Actor does

The Actor resolves each hotel URL, requests that hotel's review list, and saves reviews in batches. It follows the order and text filter you select. If a request fails, it retries with another request profile before moving to the next hotel.

Reviews that contain only a score and no text or title are skipped. Every saved record keeps the hotel, reviewer, stay, and response metadata that Booking.com returns.

Input

Field	Type	Default	Purpose
startUrls	Array of strings	The Savoy hotel page	Booking.com hotel page URLs in the /hotel/{country}/{slug}.html form.
results_wanted	Integer	20	Maximum number of unique reviews to save across all hotel URLs.
searchText	String	Empty	Keep reviews that contain this phrase. Leave empty for all reviews.
sortBy	String	MOST_RELEVANT	Review order. Options: MOST_RELEVANT, NEWEST_FIRST, OLDEST_FIRST, SCORE_DESC, SCORE_ASC.
proxyConfiguration	Object	None	Apify Proxy or your own proxy URLs. Use it when Booking.com blocks the run IP.

URLs may include a language suffix, a query string, or a reviews hash. The Actor reads the country and hotel slug from the path and ignores tracking parameters.

Output

Field	Description
hotelName	Hotel name from the hotel page data.
hotelUrl	Canonical hotel URL with the reviews tab.
reviewId	Booking.com review hash.
reviewUrl	Review page URL with the review hash.
reviewTitle	Review headline.
reviewText	Positive and negative text combined.
positive	Positive review text.
negative	Negative review text.
rating	Review score.
publishedAt	Publication date in YYYY-MM-DD form.
language	Review language code.
stayDate	Stay month in YYYY-MM form.
roomType	Room name from the booking.
travelerType	Guest type, such as Couple or Family.
reviewerName	Reviewer display name.
reviewerCountry	Reviewer country name.
reviewerCountryCode	Reviewer country code.
reviewerAvatarColor	Avatar color supplied by Booking.com.
reviewerAvatarUrl	Avatar image URL.
reviewerAnonymous	Whether the reviewer is anonymous.
reviewerShowCountryFlag	Whether the country flag is shown.
helpfulVotesCount	Number of helpful votes.
isApproved	Whether Booking.com approved the review.
isTranslatable	Whether the review can be translated.
textTrivialFlag	Booking.com text flag for short or trivial reviews.
partnerReply	Hotel reply, when one exists.
checkinDate	Stay check-in date.
checkoutDate	Stay check-out date.
numNights	Number of nights in the stay.
stayStatus	Booking.com stay status.
roomTypeId	Room type identifier.
customerType	Booking.com customer type.
photos	Review photos with size and URL entries.
photoCount	Number of review photos.

Null and empty values are removed from each record.

Example output

The sample below is trimmed to the main fields.

{
    "hotelName": "The Savoy",
    "hotelUrl": "https://www.booking.com/hotel/gb/the-savoy.en-gb.html#tab-reviews",
    "reviewId": "907d8c1fcec3cc9f",
    "reviewUrl": "https://www.booking.com/reviews/gb/hotel/the-savoy.en-gb.html#review_907d8c1fcec3cc9f",
    "reviewTitle": "We had a terrific time in the Savoy!",
    "reviewText": "Positive: Super friendly staff!",
    "positive": "Super friendly staff!",
    "rating": 10,
    "publishedAt": "2026-09-17",
    "language": "en",
    "stayDate": "2026-09",
    "roomType": "Superior Queen Room",
    "travelerType": "Couple",
    "reviewerName": "Bert",
    "reviewerCountry": "Netherlands",
    "reviewerCountryCode": "nl",
    "reviewerAnonymous": false,
    "helpfulVotesCount": 0,
    "isApproved": true,
    "isTranslatable": false,
    "numNights": 2
}

Reviews across several hotels

Add URLs in the order you want them processed. The Actor keeps fetching from each hotel until it reaches results_wanted or runs out of matching reviews.

{
    "startUrls": [
        "https://www.booking.com/hotel/gb/the-savoy.en-gb.html",
        "https://www.booking.com/hotel/gb/hampton-by-hilton-london-city.en-gb.html"
    ],
    "results_wanted": 40,
    "sortBy": "NEWEST_FIRST"
}

Text filter and order

searchText uses the review text filter on Booking.com. For example, breakfast returns reviews that mention breakfast, and the dataset text is saved without highlight markup.

sortBy applies to every hotel in the run:

• MOST_RELEVANT uses Booking.com's default order.
• NEWEST_FIRST and OLDEST_FIRST sort by publication date.
• SCORE_DESC and SCORE_ASC sort by review score.

When a request fails

The Actor rotates request profiles, retries temporary network and server failures, and keeps a small fallback set of request formats. If one hotel cannot be resolved, it logs a warning and moves to the next URL. A run fails only when no hotel returns review data.

On Apify, when Booking.com blocks the run IP, the Actor switches to Apify residential proxy automatically. If your account cannot use residential proxy, it tries the standard Apify proxy and then stops with an error. Apify Proxy usage is billed by Apify according to your plan.

If you maintain the Actor, read API_DISCOVERY.md before changing the request. It records the endpoint, required fields, request profiles, and the checks used to verify them.

FAQ

How many reviews can I collect?

The limit is results_wanted. The actual count depends on how many matching reviews the hotel has and how many of them include text or a title.

Why is the dataset smaller than results_wanted?

The hotel may have fewer matching reviews, or the text filter may exclude short score-only entries. In that case, the Actor moves to the next hotel URL.

Can I collect reviews without a text filter?

Yes. Leave searchText empty to collect all reviews in the selected order.

Do I need a proxy?

On Apify, no. The Actor switches to Apify residential proxy when the direct IP is blocked. Add your own settings under proxyConfiguration if you want to use a specific proxy group or your own proxy URLs.

Does the Actor save reviews as it goes?

Yes. Each page of reviews is saved as soon as it is validated, so a later failure does not discard earlier results.

Legal

Review data belongs to Booking.com and the people who wrote it. Use the output according to Booking.com's terms and applicable law.