Booking.com Hotel Reviews Scraper

Extract guest reviews from any Booking.com hotel page — reviewer name, score, review title, positive and negative text, room type, trip purpose, and date — all in a clean structured dataset ready for export or analysis.

Pass one URL or a hundred. The actor handles Booking.com's bot protection automatically and exits cleanly with your data.

---

Features

• Bot protection handled — runs inside a real virtual-display browser (camoufox + Xvfb) that passes Booking.com's AWS WAF challenge without any manual steps
• Full review fields — name, country, score out of 10, review title, positive text, negative text, room booked, trip purpose, and stay date, all normalised
• Multi-hotel batching — supply a list of hotel page URLs and the actor scrapes them all in one run
• Residential proxy built in — no proxy configuration needed; Apify's residential pool is used automatically
• Cheap and fast — each hotel typically completes in under 60 seconds

---

How to use

1. Open any hotel page on Booking.com — for example: https://www.booking.com/hotel/us/hilton-new-york.html
2. Paste the URL into the Hotel URLs field
3. Choose how many reviews you want, what order to sort them in, and which language
4. Click Start and your dataset will be ready in under a minute

The URL format https://www.booking.com/hotel/{country}/{hotel-name}.html works for every country and property on the platform.

---

Input

Field	Type	Default	Description
hotelUrls	string[]	(required)	One or more Booking.com hotel page URLs
maxReviewsPerHotel	integer	50	Maximum number of reviews to return per hotel (1–1000)
sortBy	string	recent	Sort order: recent · highest · lowest
language	string	en-gb	BCP 47 language filter — e.g. en-gb, de, fr. Leave blank for all languages.

---

Output

Each row in the dataset is one guest review:

{
  "hotel_name": "New York Hilton Midtown",
  "hotel_url": "https://www.booking.com/hotel/us/hilton-new-york.html",
  "hotel_slug": "hilton-new-york",
  "hotel_country": "us",
  "reviewer_name": "Julia",
  "reviewer_country": "us",
  "score": 9,
  "review_title": "Comfort and convenience in the heart of midtown.",
  "positive": "Location, comfortable beds, great shower, restaurant credit",
  "negative": "Elevator were slow and seemed unresponsive",
  "review_date": "2026-04-26",
  "stay_info": "Leisure",
  "room_type": "Urban King Room with Sofa Bed",
  "helpful_count": null,
  "owner_response": null
}

Output fields

Field	Description
hotel_name	Full hotel name as shown on Booking.com
hotel_url	Canonical hotel page URL (without language suffix)
hotel_slug	URL slug used by Booking.com for this property
hotel_country	Two-letter country code from the hotel URL
reviewer_name	Guest's display name
reviewer_country	Guest's home country code
score	Review score out of 10
review_title	Review headline written by the guest
positive	What the guest liked
negative	What the guest didn't like
review_date	Date the review was submitted (YYYY-MM-DD)
stay_info	Trip purpose — e.g. Leisure, Business, Couple, Family
room_type	Room category the guest booked
helpful_count	Number of "helpful" votes (when available)
owner_response	Management response to the review (when present)

---

Common use cases

• Sentiment analysis — process positive/negative text in bulk to surface what guests love or dislike about a property
• Competitor intelligence — scrape reviews of competing hotels to benchmark service quality, room satisfaction, and pricing perception
• Market research — aggregate reviews across a destination to understand traveller priorities and seasonal patterns
• Review monitoring — pull recent reviews for a portfolio of properties on a schedule

---

Notes

• Review volume per hotel: Booking.com's server-side render includes approximately 10 featured reviews per hotel page — their curated selection of the most recent, highest-quality reviews. These are the same reviews shown prominently to new visitors making a booking decision.
• Multiple hotels: Chain the actor with a search or list to scrape dozens of properties in one run; each runs independently so a single failure doesn't stop the batch.
• Pricing: Pay per review extracted at $1.00 per 1,000 reviews. A typical 10-review hotel run costs under $0.01.