Reviews & changes · Hospitality reviews
Hotel and restaurant review monitoring across markets and languages
A property's reputation is spread across several review sites and a dozen languages, and each site shows a visitor from each market a different slice of it. Collect the public review pages per site, language and market on a schedule, and see a complaint theme in German before it reaches your English dashboards.
Pay per GB · No subscription · Prices in EUR, excl. VAT
- €0.75
- per GB on ARC Basic, excl. VAT
- −21%
- at 1 TB (€0.59 / GB)
- 185+
- countries, city targeting
- Never
- expires — no subscription
Written for
- Hotel group revenue and guest-experience teams
- Reputation management platforms
- Restaurant group operations
- Destination marketing analysts
The short answer
- Review sites sort, filter and translate reviews according to the visitor's language and market; collect each from the market it is written for.
- Keep one sticky session per property per pass, so pagination and sort order stay coherent while you walk the review list.
- Store review IDs, dates, language and the site's own labels — verification badges, owner responses — not just the text.
- EU rules require traders that publish reviews to say whether and how they check them, and the EU, UK and US all ban fake reviews. A dated archive of the public pages is your evidence of what changed.
The problem
What goes wrong without the right IPs
The review list is sorted for someone else
Default sort, featured reviews and translations follow the visitor's language and location. Collected from one office, every market's review page looks like your home market's.
Pagination breaks when the exit changes
Review lists are paginated with state. If the exit IP changes on page three, some sites re-sort or reset, and you collect duplicates and gaps instead of a list.
Removed reviews leave no trace
Reviews are removed by moderation, by the reviewer, or after a dispute. Without your own dated snapshot, you cannot show that a review existed or when it went.
Ratings are compared across different scales
Sites use five-point, ten-point and bubble scales and weight recent reviews differently. A dashboard that averages raw scores across sites is averaging units.
Hundreds of properties, every day
A hotel group checking 300 properties on four sites in three languages makes thousands of small requests a day. From one address that is a pattern; across a pool it is a cadence.
What to collect
What you can collect
- Property and site identifiers
- One property has a different ID on each review site; the mapping is your master data.
- Review ID, date and stay or visit date
- IDs make appending idempotent; the stay date separates old experiences from new posts.
- Rating and scale
- Store the native score and the scale so you can normalise later, not at collection.
- Original language and text
- Themes are found in the original; translations are for reading, not for counting.
- Traveller type and origin market, where shown
- Business, family or couple — and the reviewer's country — explain why one market complains and another does not.
- Verification label
- Whether the site marks the review as from a verified stay or booking.
- Owner response and response date
- Response rate and speed per property and per language are standard reputation KPIs.
- Overall score, review count and rank in area
- The headline numbers a booker sees, stored as a daily series per site and market.
How it runs
How a collection run works
- 01
Map properties to sites and markets
For each hotel, restaurant or attraction, record its URL on each review site and the languages and markets you want to see it from — usually the property's own market plus its top source markets.
- 02
Walk the review list on one exit
Open a sticky session per property, site and market, sort by newest, and page until you reach a review you already have. Ten minutes is ample for most properties.
Newest-first, stop at the last known review18 linesimport requestsGATEWAY = "proxy.arcproxies.io:8000"PASSWORD = "PASSWORD"def review_session(property_id: str, country: str) -> requests.Session: user = f"USERNAME_pool-arcfast_country-{country}_session-{property_id}{country}_sesstime-10" proxy = f"http://{user}:{PASSWORD}@{GATEWAY}" s = requests.Session() s.proxies = {"http": proxy, "https": proxy} return sdef new_reviews(s, pages, known_ids: set[str]): for page in pages: # newest first for review in page(s): if review["id"] in known_ids: return # caught up yield review - 03
Snapshot the headline numbers
On the same pass, store the overall score, review count and ranking on the property page. These change daily and are what a booker reads first.
- 04
Re-check recent reviews for removal
Once a week, re-fetch reviews from the last 30–90 days by ID. A review that disappears is a data point in its own right — for disputes, for moderation patterns, and for compliance work.
- 05
Alert on themes per language
Classify themes in the original language — cleanliness, noise, staff, breakfast — and alert when one rises in one market. A German-language spike about the same room type is an operations ticket, not a reputation trend.
Setup
Recommended setup
Review pages on the big travel sites are stricter than a hotel's own website, so residential is the default and ISP takes the light, owned pages.
| Plan | Role | Session | Targeting | Price |
|---|---|---|---|---|
ARC Fast Rotating residential | Default for review sites Residential exits in each source market, held long enough to walk the review list in one coherent sort order. | Sticky, 10 minutes per property pass | Country + city | €2.00 / GB €1.12 at 1 TB (−44%) |
ARC Basic Rotating ISP | For owned and light pages Your own website's review widgets, local directories and light listing pages. Collect them on the cheapest pool. | Rotating | Country | €0.75 / GB €0.59 at 1 TB (−21%) |
ARC Ultra Premium residential | For the strictest review sites Premium residential for the sites where ARC Fast leaves you with failures you cannot otherwise explain. | Sticky, 10 minutes per property pass | Country + city | €3.50 / GB €2.17 at 1 TB (−38%) |
Prices are list prices in EUR, excl. VAT. Volume discounts apply automatically at checkout — full ladder on the pricing page.
Connect
One connection string per market
The country, city and sticky session are tokens in the username. Change the token, change the market; keep the session handle, keep the IP.
http://USERNAME_pool-arcfast_country-de_session-k3f9a1x2_sesstime-10:PASSWORD@proxy.arcproxies.io:8000Swap _country-de for any market you quote. The full grammar is on the residential proxies page.
Illustration
What a 300-property hotel group costs
An illustration with round numbers — your pages, payloads and retry rate will differ. The arithmetic is the real price ladder.
Assumptions
- Properties
- 300
- Review sites
- 4
- Markets per property
- 3
- Pages per pass
- 2 (property page + newest reviews)
- Cadence
- once a day
- Weight per page, images blocked
- 400 KB
- Bandwidth per month
- ≈ 87 GB
On ARC Fast
€128.76per month, excl. VAT
87 GB bought in one purchase at €1.48 / GB (−26% volume rung, applied automatically).
The same 87 GB on ARC Basic: €60.90 at €0.70 / GB. Move down as soon as the cheaper pool holds up.
Most days a property has no new reviews, so two pages per pass is typical; busy city hotels page deeper. The weekly removal re-check adds a modest amount on top.
Bandwidth never expires, so a bigger single purchase is only ever cheaper per GB. See the ladder.
In depth
Hospitality reviews, in depth
Why hotel reviews look different from each market
Review sites are built for travellers, so they adapt the review list to the reader: reviews in the reader's language first, translations offered or shown by default, and sort orders that favour recent and relevant reviews. A guest-experience manager in Munich and a booker in São Paulo see different first pages for the same hotel.
That is why a single global feed misleads. The complaint a Japanese guest writes about bathroom layout or a German guest writes about breakfast hours rarely shows up on the English first page — but it is exactly what that market's next booker reads.
| Signal | Varies by | How to collect it |
|---|---|---|
| Review language and order | Visitor language and market | One pass per source market, sorted newest-first |
| Overall score and review count | Site, sometimes language filter | Daily snapshot per site and market |
| Rank in area | Site, location, sometimes visitor market | Daily snapshot from an exit in the property's city or country |
| Owner responses | Property and language | Collected with each review |
What review rules now require — and why your snapshots matter
Review integrity is regulated in all three of the large markets most hotel groups sell in. In the EU, the Omnibus Directive made it material information whether and how a trader that provides access to reviews ensures they come from consumers who actually used the product, and blacklisted both claiming reviews are genuine without reasonable checks and submitting or commissioning fake reviews.
In the UK, the CMA's guidance on fake reviews (4 April 2025) explains the prohibition under the Digital Markets, Competition and Consumers Act 2024, including the obligation on traders that publish consumer reviews to take reasonable and proportionate steps to prevent and remove banned ones. In the US, the FTC announced a final rule on 14 August 2024 that bans fake reviews, review suppression through threats or intimidation, and company-controlled review sites presented as independent.
Platforms report the scale themselves. Tripadvisor's 2025 Transparency Report says it received 31.1 million reviews in 2024 and stopped 2.7 million fraudulent ones, with review boosting by businesses and their affiliates accounting for 54% of the fraud. A dated, per-market archive of the public pages is what lets you show which reviews appeared, which disappeared, and when — useful evidence in a dispute, and never a substitute for legal advice.
Hotels, restaurants and attractions are different jobs
- Hotels — several large review and booking sites, many source markets, reviews tied to stays. The cross-market work above is the core of it, and it pairs naturally with hotel rate shopping: a falling score and a falling rate in the same market are usually one story.
- Restaurants — fewer markets but more sites, including local directories and map listings. Volume per location is lower; cadence can be daily without much bandwidth.
- Attractions and tours — reviews concentrate on a few platforms and are highly seasonal. Monitor weekly off-season and daily in season.
Normalising scores across review sites
Store the native score and its scale at collection, and normalise in your warehouse. A 4.5 out of 5 and an 8.9 out of 10 are not directly comparable even after rescaling, because sites weight recency and verified stays differently.
The more robust comparison is change over time per site: the 30-day moving score on each site, and the share of reviews mentioning a theme. Those move together when something real happens at the property, which is the signal operations teams need.
Collecting responsibly
Everything here is the public, logged-out review page. Do not collect reviewer account pages beyond what the review shows, and treat reviewer names as personal data in your own storage. Respect the site's robots.txt and pace each site — the responsible travel data guide covers the practical side, and our acceptable use policy sets the boundary.
Pagination is where most of the bandwidth goes. Stop at the first review you already have, and the daily pass for a quiet property is two pages. See reducing proxy bandwidth for the rest.
Works with anything that takes a proxy URL
- Python requests / httpx
Playwright- Scrapy
Crawlee- changedetection.io
- Docker
FAQ
Questions
Something not covered? Write to contact@arcproxies.io.
Why monitor hotel reviews from different countries?
Because review sites order, filter and translate reviews for the visitor's language and market. Guests from each source market see a different first page, and complaints written in one language rarely surface on another market's page. Collecting per market shows you what each market's next guest reads.
Which review data should a hotel group store?
Per review: site, review ID, rating and scale, original language and text, review and stay date, traveller type, verification label and owner response. Per property and day: overall score, review count and rank in the area on each site and market.
Do I need a sticky session to collect hotel reviews?
It helps. Review lists are paginated and sorted with state, and changing the exit IP mid-list can re-sort or reset it. Hold one sticky session per property, site and market for the pass; on ARC Fast you set the minutes with the session time token.
Are fake reviews illegal?
In the EU, submitting or commissioning fake consumer reviews is listed as an unfair commercial practice under the Omnibus Directive. The UK prohibits fake reviews under the Digital Markets, Competition and Consumers Act 2024, and in the US the FTC's 2024 rule bans fake reviews and review suppression. This is a description of the rules, not legal advice.
How much bandwidth does hotel review monitoring use?
With images blocked and about 400 KB per page, a 300-property group checking four sites in three markets daily, two pages per pass, uses roughly 86 GB a month. Stopping pagination at the first already-known review keeps quiet properties at two pages.
Can I collect reviews that are only visible when logged in?
No. This use case covers public, logged-out review pages only. Anything behind a login, a paywall or a CAPTCHA is outside our acceptable use policy.
Sources & review
Last reviewed by the ARC Proxies team. Prices are read from the live catalog at build time.
- [1]EUR-Lex: Directive (EU) 2019/2161 (Omnibus Directive), Article 3: consumer reviews, Article 7(6) and Annex I points 23b–23c of Directive 2005/29/EC
- [2]Competition and Markets Authority: Fake reviews: guidance on the prohibition under paragraph 13 of Schedule 20 to the DMCC Act 2024 (CMA208)
- [3]Federal Trade Commission: Federal Trade Commission announces final rule banning fake reviews and testimonials
- [4]Tripadvisor: Tripadvisor's 2025 Transparency Report reveals strong review submissions and improved fraud detection
Reviews & changes
More on reviews & changes
Get started
Buy a gigabyte and try it on your own targets.
No subscription, no sales call. Code WELCOME15 takes 15% off your first purchase, and whatever you do not use is still there next quarter.