Airbnb Reviews Scraper — Guest Reviews, Ratings & Replies

An airbnb reviews scraper that returns every guest review a listing has, as structured rows: airbnb ratings, the review exactly as the guest wrote it, Airbnb's translation, the language it was written in, the reviewer, the date, the stay type and the host's public reply. It is a scrape airbnb reviews tool for hosts and analysts who want airbnb review data in bulk, not one page at a time.

Point it at listing ids, Airbnb URLs, or just a place name — a place harvests airbnb listing reviews for every listing it returns. No login, no cookies, no airbnb review api key, no browser. You pay per review row, and listings with no reviews cost nothing.

Last verified working: 2026-08-28.

What does the Airbnb reviews scraper do?

You give it listings. It gives you their airbnb guest reviews as rows you can sort, filter and export.

• review rows — rating, text (original), text_localized (Airbnb's translation into your language), language, created_at, date_label, reviewer_name, reviewer_id, reviewer_location, reviewer_is_superhost, host_reply, host_reply_date, highlight and stay_type.
• listing rows — free context per listing: title, URL, how many reviews it has and how many were taken.
• status rows — free: per listing or place, what was harvested and why anything failed.

Reviews come back in the language they were written in, with the translation alongside. A Lisbon listing in testing returned Portuguese, Spanish, German, Korean and English reviews in one run — you keep the original and you do not have to translate it yourself.

Scrape Airbnb reviews three ways

Most tools make you find listing ids first. This one does not:

• listingIds — numeric ids like 11562247, the fastest input when you already know the listings.
• startUrls — listing URLs (/rooms/12345678) or search URLs (/s/Lisbon--Portugal/homes). A search URL harvests every listing it returns.
• locationQueries — a place typed as you would on Airbnb: Lisbon, Portugal, Algarve. This is how you get airbnb reviews for a city without assembling a list by hand.

Give a place and a listing that sits inside it and they de-duplicate: the listing is harvested once, and you are billed once.

Airbnb negative reviews without paying for the happy ones

minRating is the feature that changes the economics. Set it to 3 and only reviews rated 3 or below are delivered — and only those are billed. The five-star reviews are filtered before you pay for them.

In a verified run, a listing with 441 reviews returned 4 rows under minRating: 3. That is $0.02 instead of $2.20 for the same question: what are guests actually complaining about. For review monitoring and guest feedback data on a portfolio, this is usually the only setting that matters.

Airbnb review data for hosts, managers and analysts

• Reputation monitoring. Schedule a daily run ordered by most_recent and every new review lands with its rating and text. Nothing to poll, nothing to diff.
• Competitor research. Point a place at it and pull the airbnb competitor reviews corpus for a whole market — what guests praise, what they complain about, which hosts bother to reply.
• Airbnb sentiment analysis. A rated, dated, multi-language corpus with the original text intact is exactly the shape sentiment and topic models want. language lets you split or route by language before you spend tokens on it.
• Airbnb property management reviews. Reviews carry host_reply and host_reply_date, so you can find listings where guests were left unanswered.
• Airbnb host reviews. reviewer_is_superhost flags reviewers who host themselves, which is a useful signal when you are weighing whose opinion carries weight in a market.
• Vacation rental reviews and short term rental reviews benchmarking across a market, rather than one listing at a time.

Input

Field	What it does
listingIds	Numeric Airbnb listing ids, e.g. 11562247.
startUrls	Listing URLs (/rooms/12345678) or search URLs (/s/Lisbon--Portugal/homes).
locationQueries	Places as you would type them on Airbnb: Lisbon, Portugal, Algarve.
maxListingsPerQuery	How many listings to take from each place before harvesting their reviews.
maxReviewsPerListing	0 = every review the listing has (Airbnb pages them 50 at a time). N = the first N in your chosen order. The main cost control.
reviewsSort	most_recent, highest_rated, lowest_rated, most_relevant.
minRating	0 = keep everything. 1–5 = keep only reviews at or below that rating. Filtered reviews are never charged.
locale	Language for the translations (en, es, fr, de…). The original text always comes too.
includeListingRow	Emit the free per-listing context row.
sessions, perIp	Parallel proxy sessions and the pace of each. Leave alone unless a run is throttled.
proxyConfiguration	Apify Proxy. Required — Airbnb blocks datacentre traffic that is not rotated.

Example: a scheduled complaints feed across a portfolio

{ "listingIds": ["11562247", "45307024"], "maxReviewsPerListing": 0, "reviewsSort": "lowest_rated", "minRating": 3 }

Example: the review corpus for a whole market

{ "locationQueries": ["Lisbon, Portugal"], "maxListingsPerQuery": 50, "maxReviewsPerListing": 100 }

Example: newest reviews only, from a search you already built

{ "startUrls": ["https://www.airbnb.com/s/Lisbon--Portugal/homes?room_types%5B%5D=Entire%20home%2Fapt"], "maxReviewsPerListing": 20 }

Output

{
  "type": "review", "id": "11562247", "url": "https://www.airbnb.com/rooms/11562247",
  "name": "Charming studio in Baixa", "reviews_total": 441, "locale": "en",
  "review_id": "1752111840493470784", "rating": "5",
  "text": "El departamento es muy lindo. Esta todo muy bien ubicado.",
  "text_localized": "The apartment is very nice. Everything is very well located.",
  "language": "es", "created_at": "2026-08-14T10:53:27Z", "date_label": "1 week ago",
  "reviewer_name": "Luis Armando", "reviewer_id": "313623872",
  "reviewer_location": "Buenos Aires, Argentina", "reviewer_is_superhost": false,
  "host_reply": null, "host_reply_date": null, "highlight": null, "stay_type": null
}

The free listing row gives you context, and status rows tell you what happened:

{ "type": "listing", "id": "11562247", "url": "https://www.airbnb.com/rooms/11562247", "name": "Charming studio in Baixa", "reviews_total": 441, "reviews_fetched": 100, "locale": "en" }

reviews_total is what Airbnb reports for the listing, so you can always see whether you took all of them. Status rows carry listing_id, query, listings, reviews, filtered, duplicates, fetched_at and any error. Every row — review, listing and status — carries fetched_at, which is what turns repeated runs into a time series.

How much does it cost?

Pay per event, one event per delivered review row. Free: listing rows, status rows, reviews removed by minRating, listings with no reviews, unrecognised places, and failed jobs. A listing is harvested once per run however many inputs point at it.

A listing with 441 reviews is 441 rows if you ask for all of them — maxReviewsPerListing and minRating are there to make that a deliberate choice. Verified runs behind this listing cost $0.0002–$0.0003 each in platform compute. Exact per-row pricing is on this page's pricing tab.

Airbnb reviews scraper in Python, JavaScript, curl, n8n, Make or an AI agent

In Python:

from apify_client import ApifyClient

client = ApifyClient("<YOUR_APIFY_TOKEN>")
run = client.actor("kestrel/airbnb-reviews-scraper").call(run_input={
    "listingIds": ["11562247"], "maxReviewsPerListing": 0, "reviewsSort": "lowest_rated", "minRating": 3,
})
for row in client.dataset(run["defaultDatasetId"]).iterate_items():
    if row["type"] == "review":
        print(row["rating"], row["language"], row["text"][:80])

In JavaScript (Node):

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: '<YOUR_APIFY_TOKEN>' });
const run = await client.actor('kestrel/airbnb-reviews-scraper').call({
    locationQueries: ['Lisbon, Portugal'], maxListingsPerQuery: 20, maxReviewsPerListing: 50,
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items.filter((i) => i.type === 'review').length, 'reviews');

With curl, synchronously:

curl -X POST "https://api.apify.com/v2/acts/kestrel~airbnb-reviews-scraper/run-sync-get-dataset-items?token=<YOUR_APIFY_TOKEN>" \
  -H 'Content-Type: application/json' \
  -d '{"listingIds":["11562247"],"maxReviewsPerListing":50}'

For n8n, Make and Zapier, add the actor as a step through Apify's integration and map rating, text and created_at into your alert or sheet — a "new review below 4 stars" notification is two nodes. For AI agents, the actor is callable over Apify's MCP server, so an agent can ask for a listing's reviews and read them back directly. To get airbnb reviews csv, run it and use the dataset's Export button, or append ?format=csv to the dataset items API.

Is it legal to scrape Airbnb reviews?

This reads publicly visible listing pages through the same public API the airbnb.com front-end uses, with no login and no account. Reviews are published deliberately for anyone to read, and hiQ v. LinkedIn is the usual reference point for scraping public data in the US.

Two caveats worth taking seriously. First, reviews contain personal data — a reviewer's first name, their stated location and their words. Under GDPR that makes you a controller when you store them: have a lawful basis, keep them no longer than you need, and do not re-publish reviewer identities. This actor deliberately exposes no email, phone or full name, and there is no input that would make it. Second, Airbnb's Terms of Service discourage automated access, so this is a terms question, and the risk sits with whoever operates the run. None of this is legal advice — if your use is commercial or large, ask a lawyer.

Limits and honest notes

• rating is a string as Airbnb returns it ("5"), not a number. Cast it if you are aggregating.
• reviewer_location is self-declared and often blank — treat it as a hint, not a fact.
• Translations are Airbnb's, not ours. text is always the original; text_localized is only present when Airbnb translated it.
• An unrecognised place is reported, not guessed. Airbnb answers a nonsense place with listings from all over the world; the run detects that, marks the job unknown_area and charges nothing.
• Airbnb pages reviews 50 at a time. Deep listings take several requests; maxReviewsPerListing bounds it.
• No host contact details. Not scraped, not exposed, not an input — by design.

FAQ

Does it need an Airbnb API key, login or cookies?

No. There is no public airbnb review api to hold a key for. This uses the same public GraphQL endpoint airbnb.com's own front-end calls, with no session, and it refreshes Airbnb's query hashes automatically if they change.

Can I download Airbnb reviews as CSV or Excel?

Yes. Every run writes an Apify dataset, which exports to CSV, Excel, JSON or XML from the Console, or over the API with ?format=csv. That is the short path to download airbnb reviews for a listing or a market.

How do I get only Airbnb negative reviews?

Set minRating to the ceiling you care about — 2 for one- and two-star, 3 to include mediocre stays — and reviewsSort to lowest_rated. Reviews above the ceiling are dropped before billing, so a complaints feed on a heavily reviewed listing costs cents.

Can I get reviews for every listing in a city?

Yes, with locationQueries plus maxListingsPerQuery. Airbnb caps a single search at roughly 280 listings, so for a large city, run several searches with different filters (price bands, room types) and let the de-duplication merge them.

Are reviews translated?

Each review carries language (what it was written in) and text (the original). When Airbnb offers a translation into your locale, it arrives as text_localized. So airbnb review translation is available without you calling a translation API, and the source text is never lost.

Does it include host replies?

Yes — host_reply and host_reply_date when the host answered publicly. That is how you find listings where negative reviews were left unaddressed.

How many reviews can one listing return?

All of them. Airbnb pages 50 at a time and the run follows the pages until it has what you asked for; 300 unique reviews from a single listing was verified in testing. reviews_total tells you how many exist, so you can check you got the lot.

Can I scrape Airbnb reviews without an API?

That is exactly what this is — airbnb reviews without api access, no key to apply for and no approval process. You call an Apify actor and read rows back.

What does a run cost for bulk work?

Bulk airbnb reviews are billed per delivered row, so cost scales with data rather than with time. Filtering with minRating or capping with maxReviewsPerListing cuts it directly — those rows are never charged.

Which languages and locales are supported?

Any locale Airbnb supports for translations — en, es, fr, de, pt, it, ja, ko and the rest. Set locale and text_localized follows it. The language field is independent: it tells you what the guest actually wrote in.

Airbnb reviews by listing, by market, or on a schedule

The three input modes exist because three different jobs need different shapes.

Airbnb reviews by listing is the host's job. You own five properties, you want every new review the moment it appears, and you want the ones below four stars in front of you. Give listingIds, sort most_recent, schedule it daily. The listing row tells you reviews_total so you can see growth without counting rows.

By market is the analyst's job. You are pricing a portfolio, judging a city, or building comps. Give locationQueries with maxListingsPerQuery and you get the review corpus for a market in one run — every listing's reviews, tagged with the listing they came from.

On a schedule is where the value compounds. Airbnb shows you reviews; it does not show you change. Because every row is stamped with fetched_at, a weekly run turns into a dataset that answers questions Airbnb's own interface cannot: which listings are gaining reviews fastest, whose ratings are slipping, which complaints repeat across a market and which are one bad night.

What this does not do

It does not scrape hosts' contact details, it does not touch private messages, and it does not need or accept an Airbnb login. Reviews are the public record of a stay; that is the boundary, and the actor has no input that crosses it.

Reviews as a data product

Three things make this corpus more useful than a page of text:

• The original language is preserved. Most tools translate and discard. Keeping text alongside text_localized means you can run language-specific models, or check a translation you do not trust.
• The rating is separable from the words. rating and text are distinct fields, so you can weight them independently — a four-star review with an angry paragraph is a different signal from a four-star review with "fine".
• Replies are attached. host_reply sits on the same row as the review it answers, so response rate and response tone are one GROUP BY away.

Related scrapers

Reviews tell you how a stay went; prices and availability tell you what it costs and when it is free. These share the same row shape and billing model:

• Google Hotels Prices Scraper — hotel prices and every booking site’s rate for a stay.
• Google Flights Scraper — airfare for a route and date, with a cheapest-day sweep.
• Airbnb Scraper — Airbnb prices, availability calendars and occupancy.
• Agoda Reviews Scraper — Agoda hotel reviews with separate positives and negatives.

All of them bill per delivered row, never charge for rows a filter or a spending limit removed, and write an Apify dataset you can export to CSV, Excel or JSON.