Tripadvisor API

The Tripadvisor API turns hotels, restaurants and attractions into clean, structured JSON. Search any keyword to get places with ratings, review counts and locations, then pull the full review text for any place you found, complete with author and trip details.

Built for travel analysts, hospitality teams, review-monitoring tools and AI agents that need Tripadvisor data on demand rather than as a one-off export.

What the Tripadvisor API returns

Places (search mode)

Field	Description
title	Name of the hotel, restaurant or attraction
placeType	ACCOMMODATION, EATERY or ATTRACTION
rating	Average rating out of 5
reviewCount	How many reviews the place has
location	City and region
description	The listing's own summary blurb
placeId	Feed this back in to fetch the place's reviews
url	Link to the listing
thumbnail	Photo of the place

Reviews (reviews mode)

Field	Description
title	Headline of the review
snippet	Full review text
rating	Reviewer's rating out of 5
reviewDate	When the review was posted
tripDate	When the visit actually happened
tripType	BUSINESS, COUPLES, FAMILY, FRIENDS or SOLO
language	Language the review was written in
votes	How many readers found it helpful
authorName	Reviewer's display name
authorContributions	How many reviews the author has written
totalReviews	How many reviews exist for the place in total

Use cases

The Tripadvisor API covers the jobs people hire travel review data for:

• Reputation monitoring. Track a property's rating and review count on a schedule and catch a slide before it shows up in bookings.
• Competitor benchmarking. Pull every hotel in a city, compare ratings and review volume, and see where a property actually sits.
• Review analysis. Export the full review text with trip type and trip date, then run sentiment or theme analysis over it.
• Market research. Filter a destination to hotels, restaurants or attractions and size the category before entering it.
• Travel content. Build destination guides from real ratings rather than guesswork.
• Feeding an AI agent. Hand an assistant live Tripadvisor ratings so its answers reflect what travellers are saying now.

Input parameters

Parameter	Type	Required	Default	Description
search_mode	string	yes	search	search finds places by keyword. reviews returns the review stream for one place.
query	string	in search mode	hotels in paris	The keyword to search for, such as seafood restaurants boston. Ignored in reviews mode.
place_id	string	in reviews mode	none	Numeric place identifier. Every search result row returns it as placeId. Ignored in search mode.
place_types	array	no	all types	Keep only ACCOMMODATION, EATERY or ATTRACTION. Applied to search results.
max_results	integer	no	100	Rows to return before stopping, 1 to 2000. The Actor pages automatically.
tripadvisor_domain	string	no	www.tripadvisor.com	Regional site to query, which sets result language and currency. Ten domains available.

Example output

A place row from search mode:

{
  "resultType": "place",
  "query": "hotels in paris",
  "position": 1,
  "placeId": "143336",
  "title": "Paris Las Vegas",
  "placeType": "ACCOMMODATION",
  "rating": 3.9,
  "reviewCount": 39843,
  "location": "Las Vegas, Nevada",
  "description": "Paris Las Vegas Hotel, a Caesars Rewards Destination.",
  "url": "https://www.tripadvisor.com/Hotel_Review-g45963-d143336-Reviews-Paris_Las_Vegas.html",
  "thumbnail": "https://dynamic-media-cdn.tripadvisor.com/media/photo-o/example.jpg"
}

A review row from reviews mode:

{
  "resultType": "review",
  "placeId": "143336",
  "totalReviews": 39843,
  "position": 1,
  "reviewId": "1071371920",
  "title": "Paris in July",
  "snippet": "Loved my 5 days. Close and convenient to everything.",
  "rating": 5,
  "reviewDate": "2026-08-03",
  "language": "en",
  "tripDate": "2026-07-31",
  "tripType": "FAMILY",
  "votes": 0,
  "authorName": "Michele W",
  "authorContributions": 1
}

Every row carries resultType so a mixed run stays easy to filter. Failures arrive as a single resultType: "error" row with a readable errorMessage rather than a crash.

Two modes, one Actor

Search finds places by keyword. Give it hotels in paris, seafood restaurants boston or museums in rome and it returns places with ratings and review counts. Filter to just hotels, just restaurants or just attractions with the place type filter.

Reviews pulls the review stream for a single place. It needs a place_id, which every search result row hands you as placeId. Run search first, pick the place you care about, then run reviews with that id.

The Actor pages automatically until it hits your max_results or the source runs out, so you set one number and let it work.

How to get started

1. Open the Actor and pick a search mode.
2. In search mode, type a query. In reviews mode, paste a placeId from an earlier search run.
3. Set how many results you want and run it.

View on Apify Store

Prefer to start from working code? The example repo has a Python quick start plus MCP install walkthroughs for Claude, Cursor and ChatGPT.

Pricing

Pay per result, so a run that returns 8 rows costs less than one that returns 800. There is no per-run setup fee.

Event	When it is charged
Place returned	Once per place row in search mode
Review returned	Once per review row in reviews mode

Reviews cost more than places because the source returns far fewer reviews per request. Current per-unit prices are shown on the Actor's Store card and in the Console before you run.

🔌 Use this API from Claude (MCP)

Add this Actor as a tool in Claude Code (free trial), Claude Cowork (free trial), or any other MCP client, via the hosted Apify MCP server. Use this Actor-specific URL:

https://mcp.apify.com/?tools=actors,docs,johnvc/tripadvisor-api

Setup walkthrough:

Apify MCP integration docs: https://docs.apify.com/platform/integrations/mcp

💸 Pay per run with crypto (x402)

The Tripadvisor API supports agentic payments via the x402 protocol. AI agents and MCP clients can pay for runs in USDC (on Base) with no Apify account or API token needed: point your agent at the Apify MCP server and it can discover, pay for, and run this Actor autonomously. Read the Apify x402 announcement for details.

🔌 Integrations: put Tripadvisor data on a schedule

Most people start with a one-off run and end up wanting the same query every week. That is what this section is for.

Tasks and Schedules first. Save a run configuration as a Task, then attach a Schedule to it. A nightly rating check on twenty competitor hotels costs almost nothing and gives you a trend line instead of a snapshot. This is the highest-value setup and it needs no code.

n8n, Make and Zapier. Trigger the Actor from a workflow, then route rows onward: a rating drop below a threshold into Slack, new reviews into a spreadsheet, or the full review stream into a sentiment step.

Storage. Push results straight into Supabase, Postgres or Google Sheets through the Apify integrations tab, so history accumulates rather than being overwritten each run.

MCP. Point an AI assistant at the Actor and let it fetch live review data mid-conversation. See the MCP section above for the server URL.

Webhooks. Fire an HTTP callback when a run finishes so your own service can pick up the dataset without polling.

🔗 Related Tools

Pair this with the rest of the local and reviews cluster:

• Yelp Search API for United States local business search
• Yelp Reviews API for Yelp review text
• OpenTable Reviews API for restaurant reservations and reviews
• Google Hotels API for hotel rates and availability

Older alternatives such as reviewly/tripadvisor-reviews-scraper exist, but it carries a 3-star rating and around 20 users, and returns reviews only. This API is actively maintained, covers place search as well as reviews, and returns both in one consistent schema.

FAQ

How do I get reviews for a specific hotel or restaurant?

Run the Actor in search mode first with a query that finds the place, for example paris las vegas hotel. You can also read the id straight off a Tripadvisor listing URL: it is the number after the -d. Every result row includes a placeId. Copy that value, switch the search mode to reviews, paste it into the Place ID field and run again.

Why does my search return places in the wrong city?

The search matches place names as well as locations, so a query like restaurants in chicago can surface a restaurant named "Chicago Pizza" that sits elsewhere. Add the region to the query, or filter the output on the location field.

Can I get a place's address, phone number or opening hours?

Not in this version. The upstream place-detail endpoint is currently unreliable, so this Actor ships the two modes that return data dependably rather than a third that would fail. Ratings, review counts, location and the full review stream are all available today.

How many reviews can I pull for one place?

As many as the place has. The totalReviews field on every review row tells you the size of the pool, so you can set max_results accordingly. Reviews arrive 10 per request, so large pulls take proportionally longer.

Does it work outside the United States?

Yes. Pick a regional domain in the input to control the language and currency of the results. The United Kingdom, Canada, Australia, India, Germany, France, Spain, Italy and Brazil sites are all available.

Is this a Tripadvisor API or a web scraper?

You use it as an API: structured JSON in, structured JSON out, with a documented schema and pay-per-result billing. Under the hood it reads public Tripadvisor listing pages, so it is not an official product and is not affiliated with, endorsed by, or connected to Tripadvisor. Anything visible to a logged-out visitor is available; anything behind a login is not.

Can I use this as a Tripadvisor scraper for bulk exports?

Yes. Set max_results high and the Actor pages until it hits your number or the source runs out. Because billing is per result, a bulk export costs in proportion to what you actually receive. For recurring exports, pair a Task with a Schedule as described in the Integrations section above.

How do I call this from an AI agent using MCP?

Point any Model Context Protocol client at the server URL in the MCP section above. The agent discovers the Actor as a tool, reads the input schema itself, and can fetch live ratings and reviews during a conversation without you writing glue code.

How do I run this on a schedule?

Save your configuration as a Task, then attach a Schedule to it. The Integrations section above covers this and the downstream options (n8n, Make, Zapier, Supabase, webhooks) in more detail.

What other tools work well alongside this one?

The Related Tools section above lists the rest of the local and reviews cluster: Yelp for United States business search, OpenTable for restaurant reservations, and Google Hotels for rates and availability. Running Tripadvisor and Yelp together gives noticeably better coverage of a city than either alone.

What happens if the source is temporarily unavailable?

The run finishes cleanly rather than crashing. If some rows were already returned, they are kept and the status message explains where it stopped. If nothing was returned, you get a single error row with a readable explanation and you are not charged for results you did not receive.

Agent Skills

Ready-made skills for Claude Code (free trial) and other agents that drive this Actor. Install with npx skills add johnisanerd/<name>.

• claude-skill-tripadvisor-scraper: Tripadvisor places and reviews as structured JSON
• claude-skill-tripadvisor-hotel-reviews: a hotel reviews dataset for sentiment and monitoring

Featured Tasks

Ready-to-run examples on the Apify Store.

• Tripadvisor Reviews API for Hotels and Restaurants
• Export Tripadvisor Reviews to CSV or JSON
• Fresh Tripadvisor Hotel Reviews Dataset on Demand
• Scrape Tripadvisor Reviews in Python
• Tripadvisor Reviews in Claude via MCP

---

🌐 About Alpha OSINT

This Actor is part of Alpha OSINT, toolset of financial and operations data sources and APIs. For support or requests for this actor, please start a ticket directly on our support page.

Last Updated: 2026.08.06