Tripadvisor Scraper — Restaurants, Hotels & Attractions

Paste a Tripadvisor listing URL. Get every place on it as clean, structured data — including the fields the page never shows you, like GPS coordinates and the exact review count.

Works on every Tripadvisor country domain: .com, .es, .co.uk, .com.ar, .com.mx, .com.br, .fr, .de, .it.

---

What you get per place

Field	Example
name	Rosi La Loca
category	restaurant · hotel · attraction
rating	4.6
reviewsCount	9183
rank	position in the Tripadvisor ranking
priceRange	$$ - $$$
categories	["International", "Spanish"]
contact.phones	["+34915326681"] — normalised to E.164
location	street, city, postcode, latitude and longitude
openingHours	one entry per day, HH:MM
images	full-resolution URLs

{
  "id": "a1b2c3d4e5f6a7b8",
  "url": "https://www.tripadvisor.com/Restaurant_Review-g187514-d7204621-Reviews-Rosi_La_Loca-Madrid.html",
  "source": "tripadvisor",
  "name": "Rosi La Loca",
  "category": "restaurant",
  "rating": 4.6,
  "reviewsCount": 9183,
  "rank": 1,
  "priceRange": "$$ - $$$",
  "categories": ["International", "Spanish"],
  "contact": { "phones": ["+34915326681"] },
  "location": {
    "address": "Calle De Cadiz, 4, 28012 Madrid Spain",
    "postalCode": "28012",
    "latitude": 40.415836,
    "longitude": -3.702976
  },
  "listingId": "7204621"
}

How to use it

Scrape a city's restaurants

Open Tripadvisor, filter however you like, copy the URL:

{
  "startUrls": [{ "url": "https://www.tripadvisor.com/Restaurants-g187514-Madrid.html" }],
  "maxItems": 500
}

Build a lead list of well-rated venues with phone numbers

{
  "startUrls": [{ "url": "https://www.tripadvisor.com/Restaurants-g187514-Madrid.html" }],
  "minRating": 4,
  "minReviews": 100,
  "onlyWithPhone": true
}

Monitor new listings

Schedule it with includeSeen off — each run returns only places that appeared since the last one.

Why paste a URL instead of typing a city?

Tripadvisor URLs contain an internal geo id — g187514 is Madrid — that cannot be derived from a city name without an extra lookup. An Actor that accepts "Madrid" has to guess that id, and a wrong guess silently returns another city's listings.

Pasting the URL is slightly less convenient and completely unambiguous. You get the city you asked for, every time.

Input reference

Field	Type	Default	What it does
startUrls	array	example URL	Tripadvisor listing pages
category	enum	any	Keep only restaurants, hotels or attractions
nameKeywords	array	[]	Keep places whose name matches any word
minRating	integer	—	Minimum rating, 0–5
minReviews	integer	—	Minimum review count
onlyWithPhone	boolean	false	Keep only places with a phone number
maxItems	integer	200	Hard stop. Your cost control
maxPagesPerUrl	integer	10	Listing pages per URL (~30 places each)
includeSeen	boolean	true	false returns only places new since the last run
proxyTier	enum	datacenter	Starts cheap, escalates only if blocked
maxProxyTier	enum	residential	Set to datacenter to never spend residential bandwidth
maxRequestsPerMinute	integer	60	Throttle. Often more effective than switching proxies

Cost

Charged per place saved. Places removed by your filters or already seen are not charged.

The Actor starts on datacenter proxies and escalates to residential only when it detects an anti-bot challenge, so you do not pay for residential bandwidth unless the site actually demands it.

FAQ

Which URLs work? Any Tripadvisor listing page — restaurants, hotels or attractions — on any country domain.

Why do some places have no phone or no coordinates? Because Tripadvisor does not publish them for that venue. The field comes back null rather than being guessed.

Why fewer results than the page shows? Almost always maxItems or maxPagesPerUrl. Raise them. The run log reports how many places each filter removed.

Is this legal? Only public listing data is extracted — the same information any visitor sees without logging in. No login is bypassed and no personal data is collected.

Support

Found a listing URL that does not parse? Open an issue with the URL. Reports with a URL get fixed fastest.