Google Hotels Scraper

Google Hotels Scraper extracts hotels from Google Hotels for any place and any dates — the nightly price and the stay total with taxes, the guest rating and review count, star class, property type, amenities, coordinates, photos, check-in and check-out times, the Google Hotels page and the booking link — and, on request, every booking site's offer for a hotel, its full address and phone, and the guest reviews Google aggregates from Tripadvisor, Trip.com and the booking sites. Google shows 20 hotels and stops; this Actor walks Google's own sort orders, star classes and price bands and de-duplicates by hotel, so one search for a city returns well over a hundred properties instead of twenty. Pure HTTP, no browser, no login, no API key — a search of 150 hotels takes about 17 seconds.

Paste a city, pick dates, click Start, and download JSON, CSV or Excel — or pull the rows through the Apify API into Google Sheets, n8n, Make, Zapier, a database or an AI agent.

What is Google Hotels Scraper?

Google Hotels is the price comparison inside Google Travel: for a place and a stay it lists the properties, the cheapest rate it found across Booking.com, Expedia, Agoda, Hotels.com, Trip.com, the hotel's own site and dozens of smaller sellers, the aggregate guest rating, and the amenities. There is no official Google Hotels API. This Actor is the Google Hotels API alternative: it reads the same JSON endpoint the Google Hotels page itself calls, plus the server-rendered hotel page, and turns them into flat rows you can sort, filter and join.

It answers the questions a revenue manager, an OTA analyst or a travel-data team actually asks:

• What does every hotel in Lisbon cost for the first weekend of October, and how does that compare with last month?
• Which 4- and 5-star properties in Ubud are rated 4.5 or better, and what is the cheapest nightly rate among them?
• Which booking sites sell a given hotel, at what total, and where is the official site's price against Booking.com?
• What do guests praise and complain about — service, location, dining, cleanliness — and how many say so?

Three modes, one row shape:

Mode	You give	You get
Search	places, check-in, check-out, guests	hotels with price for that stay, rating, reviews, class, amenities, location, photos, booking link
Search + hotel pages (add-on)	the same, with read each hotel's page on	everything above plus every booking site's offer with a deep link, the full amenity list, address and phone, review scores per source, guest reviews
Hotel pages	Google Hotels URLs or entity ids	one row per hotel with offers, reviews and full details — no search

What data can Google Hotels Scraper extract?

Every hotel row carries these fields. Nothing is invented: a field Google did not show for a property is empty rather than guessed.

Field	Example	Notes
name	Alaya Resort Ubud
hotelClass, hotelClassLabel	5, "5-star hotel"	empty for unrated properties
propertyType	resort hotel	hotel, villa, bed and breakfast, hostel, apartment, guest house …
rating, reviewCount	4.7, 4006	Google's aggregate, 5-point scale
pricePerNight, pricePerNightDisplay	340.29, "$340"	the cheapest rate Google found for the searched stay, taxes included, averaged per night
totalPrice, totalPriceDisplay	680.59, "$681"	the whole stay
priceBeforeTaxes, taxesAndFees	562.45, 118.13	when Google itemises them
currency	USD	any code Google supports
checkIn, checkOut, nights, adults	2026-10-10, 2026-10-12, 2, 2	what the price was quoted for
description	Polished rooms in a chic getaway offering a spa…	Google's one-line summary
address, phone, website	Jl. Hanoman, Ubud…, +62 361 972200, alayahotels.com	address and phone come with the hotel page; the website usually comes with search
country, latitude, longitude	ID, -8.5188, 115.2635
checkInTime, checkOutTime	3:00 PM, 12:00 PM
amenities, amenityIds	["Free Wi-Fi", "Outdoor pool", "Spa"], [29, 22, 26]	names for the ids Google uses; the hotel page adds the full list with free/paid qualifiers
ratingHistogram	{stars5: 3272, stars4: 518, …}	review counts per star
reviewTopics	Service 705 mentions, 667 positive, 28 negative	what guests write about
nearbyPlaces	Sacred Monkey Forest Sanctuary, 7 min	landmarks, transport, airport with travel time
thumbnail, photos	image URLs
entityId, placeId, hotelUrl, bookingUrl	ChgI…, 0x2dd2…, google.com/travel/hotels/entity/…, google.com/travel/clk/…	the booking link goes to the cheapest seller for the searched stay
location, searchRank	Ubud, Bali, 1	the place you searched and the row's position
offers[]	Booking.com $264, Vio.com $166, Expedia $233 …	hotel page: seller, stay total, per night, dates quoted, deep link, official-site flag
cheapestOfferSeller, cheapestOfferTotal, cheapestOfferPerNight	Vio.com, 166.33, 166.33	the lowest of the offers
reviewSources[]	Tripadvisor 4.8 (1,490), Trip.com 4.7 (104)	rating and count per source Google aggregates
reviews[]	source, author, date, rating, text, link	the reviews Google renders on the hotel page

Beyond Google's 20 results

Google Hotels returns 20 properties per request, and its "next page" is the same request with a different filter — there is no offset to walk. Other scrapers stop at those 20, or page blindly and return the same hotels again.

This Actor reads the way Google itself lets you explore: it asks for the same place sorted by relevance, by lowest price, by highest rating and by most reviewed, then per star class, then under a few price caps, and keeps every hotel once. Measured on Ubud, Bali on 14 September 2026: 197 distinct hotels from 15 requests, against 20 from the first request. Set Hotels per place to what you need; the walk stops when the count is reached or when four facets in a row bring nothing new, so a small town does not burn requests looking for hotels it does not have.

The price on each row is still Google's own cheapest rate for your dates — the facet walk only changes which hotels come back, never what Google says they cost.

How much does it cost to scrape Google Hotels?

The Actor is priced per hotel row, with an add-on per hotel page read. There is no start fee, and rows that failed or came back empty are never charged.

Event	Price	What it buys
Hotel	$3.00 per 1,000	one hotel with price, rating, class, amenities, location, photos, booking link
Hotel page read (add-on)	$8.00 per 1,000	that hotel's offers from every booking site, full amenities, address, phone, review sources, guest reviews

Two hundred hotels for a city cost $0.60. Reading the page of the fifty you shortlist adds $0.40. There is no residential proxy and no browser to pay for: the Actor's own platform usage is a fraction of a cent per run. Bronze, Silver and Gold plans get 10 %, 20 % and 30 % off.

The free plan includes $5 of usage every month — about 1,600 hotel rows, or a full city every day of the month.

How to scrape Google Hotels

1. Open the Actor and type the place the way you would in Google Hotels: a city, a district, a landmark. One per line for several.
2. Set check-in and check-out. Leave them empty for a two-night stay 30 days out — the Actor never searches a date that has passed, even from a saved task.
3. Pick the currency and how many hotels per place you want. Twenty is one Google page; a hundred or more is what Google would never show you at once.
4. Optionally filter — star class, minimum rating, maximum price per night, pool — and turn on read each hotel's page to get every booking site's offer and the guest reviews.
5. Click Start. Results appear in the dataset as the run goes; export them as JSON, CSV or Excel, or read them through the API.

⬇️ Input

Hotels for a city and dates

{
  "locations": ["Ubud, Bali"],
  "checkIn": "2026-10-10",
  "checkOut": "2026-10-12",
  "adults": 2,
  "currency": "USD",
  "maxHotelsPerLocation": 100
}

Shortlist with filters

{
  "locations": ["Lisbon", "Porto"],
  "checkIn": "2026-11-06",
  "checkOut": "2026-11-08",
  "hotelClass": ["4", "5"],
  "minRating": "4.5",
  "maxPricePerNight": 250,
  "sortBy": "lowest_price",
  "maxHotelsPerLocation": 60
}

Filters are sent to Google where it supports them and checked again on every row, so a 3-star property cannot slip into a 4–5-star shortlist through a ranking quirk.

Offers from every booking site, plus reviews

{
  "locations": ["Shibuya, Tokyo"],
  "checkIn": "2026-12-20",
  "checkOut": "2026-12-23",
  "maxHotelsPerLocation": 30,
  "includeOffers": true,
  "maxOfferLookups": 30,
  "includeReviews": true,
  "maxReviews": 10
}

Hotel pages by URL

{
  "hotelUrls": [
    "https://www.google.com/travel/hotels/entity/ChgI56yqtNfw86kxGgwvZy8xMXpicXFianMQAQ",
    "ChkIjJ2ht8Gcu_h5Gg0vZy8xMWM1cnRwc3lmEAE"
  ],
  "includeReviews": true
}

Entity ids come from an earlier run's entityId column, so a daily price monitor for a fixed set of hotels needs no search at all.

Every input field

Field	Default	Meaning
locations	["Ubud, Bali"]	places to search, one per line
checkIn, checkOut	30 days out, 2 nights	YYYY-MM-DD
adults	2	guests in one room, 1–8
maxHotelsPerLocation	40	how deep to go per place, up to 500
currency	USD	any ISO code
sortBy	relevance	relevance, lowest_price, highest_rating, most_reviewed — the first facet read and the searchRank order
hotelClass	any	keep only these star classes
minRating	any	3.5, 4 or 4.5
maxPricePerNight	none	in the output currency
poolOnly	false	only hotels with a pool
includeOffers	false	read each hotel's page for offers, full amenities, address, phone, review sources ($)
maxOfferLookups	20	cap on pages read per run
includeReviews	false	include guest reviews from the hotel page
maxReviews	10	per hotel
hotelUrls	—	Google Hotels URLs or entity ids to read directly
language, market	en, us	language of labels; country Google prices for
proxyConfiguration	Apify datacenter	residential only if a run reports consent pages

⬆️ Output

One row per hotel. Error rows carry type: "error" and an explanation, and are never charged.

{
  "type": "hotel",
  "name": "Alaya Resort Ubud",
  "hotelClass": 5,
  "hotelClassLabel": "5-star hotel",
  "propertyType": "resort hotel",
  "rating": 4.7,
  "reviewCount": 4006,
  "pricePerNight": 340.29,
  "pricePerNightDisplay": "$340",
  "totalPrice": 680.59,
  "totalPriceDisplay": "$681",
  "priceBeforeTaxes": 562.45,
  "taxesAndFees": 118.13,
  "currency": "USD",
  "checkIn": "2026-10-10",
  "checkOut": "2026-10-12",
  "nights": 2,
  "adults": 2,
  "description": "Polished rooms in a chic getaway offering a spa, a low-key bar & a bright restaurant.",
  "website": "https://alayahotels.com/alaya-resort-ubud/",
  "country": "ID",
  "latitude": -8.5187697,
  "longitude": 115.263528,
  "checkInTime": "3:00 PM",
  "checkOutTime": "12:00 PM",
  "amenities": ["Air conditioning", "Bar", "Free breakfast", "Fitness center", "Free parking", "Pet-friendly", "Outdoor pool", "Restaurant", "Spa", "Free Wi-Fi"],
  "amenityIds": [2, 4, 6, 8, 16, 18, 22, 23, 26, 29],
  "ratingHistogram": { "stars5": 3272, "stars4": 518, "stars3": 114, "stars2": 42, "stars1": 60 },
  "reviewTopics": [
    { "topic": "Service", "mentions": 705, "positive": 667, "negative": 28, "score": 0.81 },
    { "topic": "Location", "mentions": 423, "positive": 384, "negative": 11, "score": 0.74 }
  ],
  "nearbyPlaces": [
    { "name": "Sacred Monkey Forest Sanctuary", "travelTime": "7 min" },
    { "name": "I Gusti Ngurah Rai International Airport", "travelTime": "1 hr 8 min" }
  ],
  "thumbnail": "https://lh3.googleusercontent.com/…",
  "photos": ["https://lh3.googleusercontent.com/…"],
  "entityId": "ChgI56yqtNfw86kxGgwvZy8xMXpicXFianMQAQ",
  "placeId": "0x2dd23d6c35a395c7:0x3153cf85768a9667",
  "hotelUrl": "https://www.google.com/travel/hotels/entity/ChgI56yqtNfw86kxGgwvZy8xMXpicXFianMQAQ",
  "bookingUrl": "https://www.google.com/travel/clk/hi?qid=…",
  "location": "Ubud, Bali",
  "searchRank": 1,
  "scrapedAt": "2026-09-14T13:14:01.199Z"
}

With the hotel page read

{
  "address": "Jl. Raya Sanggingan, Ubud, Kecamatan Ubud, Kabupaten Gianyar, Bali 80571, Indonesia",
  "phone": "+62 361 974330",
  "offers": [
    { "seller": "Vio.com", "totalPrice": 166.33, "totalPriceDisplay": "$166", "pricePerNight": 166.33, "checkIn": "2026-09-28", "checkOut": "2026-09-29", "nights": 1, "bookingUrl": "https://www.vio.com/…", "isOfficialSite": false },
    { "seller": "Pita Maha Resort & Spa", "totalPrice": 264.08, "totalPriceDisplay": "$264", "pricePerNight": 264.08, "checkIn": "2026-09-28", "checkOut": "2026-09-29", "nights": 1, "bookingUrl": "https://pitamaha.reserveonline.id/…", "isOfficialSite": false }
  ],
  "offersCheckIn": "2026-09-28",
  "offersCheckOut": "2026-09-29",
  "cheapestOfferSeller": "Vio.com",
  "cheapestOfferTotal": 166.33,
  "reviewSources": [
    { "source": "Tripadvisor", "rating": 4.8, "ratingScale": 5, "reviewCount": 1490 },
    { "source": "Trip.com", "rating": 4.7, "ratingScale": 5, "reviewCount": 104 }
  ],
  "reviews": [
    { "source": "Tripadvisor", "author": "N8293ABtammyh", "date": "2 weeks ago", "rating": 5, "ratingScale": 5, "text": "We had an absolutely wonderful stay…", "url": "https://www.tripadvisor.com/…" }
  ],
  "detailsFetched": true
}

One thing to know about offers, stated plainly because it is easy to misread: Google's hotel page quotes every booking site for the stay Google picks by default — about two weeks out — and ignores the dates in the address bar. The Actor reads those dates back and stamps them on each offer (offersCheckIn, offersCheckOut), so the seller list, the deep links and the relative order of sellers are exact, but the offer totals are for that default stay. The price for your searched dates is pricePerNight / totalPrice on the same row — Google's cheapest rate across the same sellers — and bookingUrl opens that seller for your dates.

The dataset has three table views: Hotels, Offers per booking site (one line per seller) and Guest reviews (one line per review).

Use cases for Google Hotels data

Rate shopping and revenue management

Run the same city every morning and diff pricePerNight per entityId: who dropped, who raised, where you sit against the compset for the next weekend. The facet walk means the compset is the whole market, not the twenty hotels Google felt like showing.

OTA and metasearch analysis

With the hotel page read, offers tells you which sellers list a property and at what total. Count how often the official site is the cheapest, which OTAs undercut whom, and where a hotel is missing from a channel entirely.

Market research and site selection

Class, price band, rating and amenity mix for every property in a district, with coordinates for a map. Compare Shibuya with Shinjuku, or Ubud with Seminyak, in one run.

Guest-experience research

reviewTopics gives you what guests talk about and how it splits positive against negative, for every hotel in a city, without reading a review. reviews gives you the text when you do want to read.

Travel products and AI agents

Flat rows with a stable entityId, a booking link and a per-night price are what a trip planner, a price-alert bot or an agent answering "find me a 4-star under $150 near the station" needs. The Actor runs from the Apify API and through the Apify MCP server.

Integrations

• API: start a run with POST https://api.apify.com/v2/acts/lergassy~google-hotels-scraper/runs?token=… and read …/datasets/{id}/items?format=csv.
• Clients: apify-client for Python and JavaScript wrap the same calls.
• Google Sheets, n8n, Make, Zapier: use the Apify integration; the flat row shape needs no transformation.
• Webhooks: fire on run finish to pull the dataset into your own system.
• MCP: the Actor is callable as a tool through the Apify MCP server, so an AI agent can search hotels and read prices in a conversation.
• Schedules: a daily task with a fixed hotelUrls list is a price monitor for your own properties and competitors.

from apify_client import ApifyClient

client = ApifyClient("<APIFY_TOKEN>")
run = client.actor("lergassy/google-hotels-scraper").call(run_input={
    "locations": ["Lisbon"], "checkIn": "2026-11-06", "checkOut": "2026-11-08",
    "hotelClass": ["4", "5"], "maxHotelsPerLocation": 100,
})
for hotel in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(hotel["name"], hotel["pricePerNight"], hotel["rating"])

🤖 For AI agents and LLM apps

Inputs that matter: locations (list of place strings), checkIn and checkOut (YYYY-MM-DD), maxHotelsPerLocation (integer), currency (ISO code), optional hotelClass (list of "1"–"5"), minRating ("3.5" | "4" | "4.5"), maxPricePerNight (number), includeOffers (boolean, costs one page read per hotel), hotelUrls (list of Google Hotels URLs or entity ids).

Output: one JSON object per hotel with name, hotelClass, rating, reviewCount, pricePerNight, totalPrice, currency, checkIn, checkOut, amenities, latitude, longitude, address, hotelUrl, bookingUrl, entityId; with includeOffers, also offers[] (seller, totalPrice, bookingUrl), reviewSources[] and reviews[]. Rows with type: "error" explain what to change. Sort or filter on the numeric fields directly; use entityId to refer to the same hotel across runs.

Limits, honestly

• Search returns what Google shows for the place. Google decides which properties belong to "Ubud, Bali"; the facet walk enlarges that set but cannot add a hotel Google does not list there.
• Offers on the hotel page are for Google's default stay, not the searched dates — see the output section. The searched stay's cheapest price is on every row regardless.
• Address and phone arrive with the hotel page; Google includes them in search results only sometimes.
• Amenities from search are the ids Google tags the property with, named where the table in this Actor knows them and returned raw otherwise; the hotel page carries the full labelled list.
• Children and rooms are not yet inputs — prices are for the given number of adults in one room.
• Rate limits. Google answers these requests from datacenter IPs today; each request retries on a new IP, and a run that meets a consent page says so in the log. Residential proxies are a switch away if that changes.

❓ FAQ

Is there a Google Hotels API?

Not a public one. This Actor reads the same endpoint the Google Hotels page uses, which is why the fields match what you see on the page — including the itemised taxes and the review topics.

How many hotels can I get for one city?

As many as Google lists for it, up to the cap you set. The facet walk measured 197 distinct hotels for Ubud from 15 requests; a big city yields several hundred. Set maxHotelsPerLocation to what you need.

Are the prices for my dates?

Yes for pricePerNight and totalPrice — those are Google's cheapest rate across booking sites for the check-in and check-out you gave, taxes included. The per-seller offers from the hotel page are for Google's default stay; the row says which dates.

Does it include taxes?

totalPrice includes taxes and fees; priceBeforeTaxes and taxesAndFees are split out when Google itemises them.

Can I get reviews?

Yes — turn on includeReviews with the hotel page add-on or with hotelUrls. You get the reviews Google renders on the page (source, author, date, rating, text, link), typically 10–20 per hotel, plus the rating and review count per source in reviewSources.

Can I use it with the Apify API or an MCP server?

Both. Start runs and read datasets through the REST API or the Python and JavaScript clients, and call the Actor as a tool from any MCP-capable agent through the Apify MCP server.

Can I integrate it with other apps?

Through the Apify integrations for Google Sheets, n8n, Make and Zapier, or through webhooks on run finish.

Why does a default run search 30 days out?

Because a saved task or a scheduled monitor must never search a date that has already passed and come back empty. Give explicit dates and they are used; give a past date and the row says the date was adjusted.

Is it legal to scrape Google Hotels?

The Actor reads public pages and public data — prices, ratings, amenities — with no login and no personal data beyond reviewer display names Google already publishes. How you use the data is subject to your local law and Google's terms; consult a lawyer for your case.

One row per hotel?

Yes. Offers and reviews are arrays on the hotel's row; the dataset views unwind them into one line per seller or per review when you want a flat table.

Your feedback

If a field is missing, a place returns fewer hotels than Google shows, or a price does not match the page, open an issue on the Actor with the input you used — it is answered quickly and fixed in the next build. A short review after your first run helps other people find a scraper that works.

You might also like

Actor	What it does
Google Flights Scraper	fares, price calendar and booking options by route and dates
Booking.com Hotel Search Scraper	Booking.com search with rooms, rates and reviews
Agoda Reviews Scraper	guest reviews and ratings from Agoda
Trip.com Scraper	Trip.com hotel prices and availability
Airbnb Scraper	vacation rental prices and availability

Also known as

Google Hotels API, Google Hotel Search scraper, Google Travel hotels scraper, hotel price scraper, hotel rate shopping tool, hotel price comparison API, scrape hotel prices from Google, Google hotel reviews scraper, hotel metasearch data.