Review sources

How to add Booking.com reviews to your hotel website

30 August 2026 · JustReview

To add Booking.com reviews to your website you connect your public property page to a review widget and paste a short snippet where the booking decision happens, usually the homepage, the rooms page or the direct booking form. There is nothing to export from the Extranet and no plugin to install, because the reviews are already public and the widget reads them from there. The step almost everyone skips is the one that matters: Booking reviews are not shaped like Google reviews, and if you drop them on your own page unread you will publish your guests’ complaints alongside their praise, in the same paragraph, under a five-star rating.

We import Booking.com on behalf of paying properties. Everything below is measured from our production database on 30 August 2026, across 3,249 Booking.com reviews belonging to 12 properties. It is a small source by our standards, which is exactly why the numbers are worth reading: nobody else has bothered to look at it closely.

A Booking.com review is two reviews stapled together

This is the finding that should change what you do this week. Booking does not ask a guest to write a review. It asks two separate questions, what did you like and what did you not like, and it publishes both answers as one review. When that review reaches your website, both halves come with it.

The scale of it is larger than anyone expects. Of our 3,249 Booking.com reviews, 1,216 carry a written complaint, which is 37.4%. Filter to only the perfect scores and the number barely moves.

Read the top row again. Roughly one in three of your five-star Booking reviews contains a sentence about what the guest disliked. Paste those on a direct booking page without looking and you have built the most persuasive argument against booking with you that your competitors never had to write.

There is a length signal too, and it is useful. A Booking review with a complaint in it averages 274 characters. One without averages 57. The negative half is where the writing happens, because a happy guest types “great location” and an irritated one explains themselves. So the reviews that look richest in a widget preview are disproportionately the ones carrying the criticism.

None of this makes Booking a bad source. It makes it a source that needs a decision from you before it goes live: show the positive half only, or moderate review by review, or accept the complaints deliberately because visible imperfection reads as honest. All three are defensible. Publishing by accident is not.

Your stars will land on halves, and that is correct

Booking scores a stay out of 10. Five-star widgets halve that number, so a 9 out of 10 becomes a 4.5. Across our Booking data, 1,014 reviews, 31.2%, sit on a half star, with 4.5 alone accounting for 840 of them.

The practical consequence is a test you can run on any Booking widget in ten seconds. If it shows you a clean five-star distribution with no half values, it is rounding, and rounding upward turns a 4.58 average into something closer to 4.8. Your real Booking average across our data is 4.58, against 4.70 for Google across the 386,461 Google reviews we hold. A widget that hides the halves is not tidying your data, it is quietly overstating your property.

Most of your guests did not write in English

Booking.com is the most linguistically scattered source we import, and it is not close. Of the 3,249 reviews, 42.8% are in a language other than English, spread across 32 detected languages. Compare that with the two sources everyone else writes about.

The distribution inside Booking is lumpy rather than even: 1,860 English, 769 Polish, 122 Spanish, 87 German, 57 French, 52 Czech, 36 Italian, then a long tail. For a property that sells to two or three markets, that is a feature, not a problem, provided your widget can filter by language per page. A German visitor reading four Czech reviews learns nothing. The same German visitor reading four German reviews of the same property is materially closer to booking. If you run language variants of your site, this is the highest-value setting on the widget, and it is the one nobody configures.

There is a compensating strength worth knowing. Only 15.6% of Booking reviews have no text at all, against 32.3% for Google. A third of your Google reviews are stars with silence attached; on Booking, guests write. That is why a Booking source works in a testimonial layout, where Google often needs a rating summary instead to avoid a page of empty cards.

Booking is an archive, not a feed

Two numbers define the rhythm of this source. The oldest Booking review we hold dates from 13 October 2019, and the deepest single property history spans 2,506 days, close to seven years. There is no twelve-month window cutting your history off. But in the 30 days to 30 August 2026, only 57 new Booking reviews arrived across the whole customer base, from 4 of the 12 properties. In the same window our largest source delivered 6,995.

So plan the way you would plan for any archive. The day you connect, years of proof appear at once and your page is full. After that it trickles, because a hotel takes a hundred bookings to earn what a marketplace seller earns in a week. If you are choosing where to put your setup effort, connect Booking for the backlog and connect something with a faster pulse for the flow. We laid out which sources still deliver reliably in which review sources still work in 2026, and the general case for moving marketplace proof onto your own domain is in marketplace reviews on your own website.

Booking alone is thin proof for a direct booking page

Of the 12 properties we import Booking for, only 2 have it as their sole source. The rest pair it with something, and the pairing is consistent: 7 also import Google, 3 import Airbnb, 2 import TripAdvisor. That is not an accident of our customer base, it is the shape of the problem. A Booking review proves the room matched the photos. It says nothing about whether you answer the phone, whether the direct rate is honoured, or whether the website the visitor is currently on is run by the same people. Those are exactly the doubts that stop a direct booking and send the guest back to the OTA, where the commission lives.

The pattern that works is Booking as the volume and depth layer, with a second source carrying the things Booking cannot say. If you have never collected reviews under your own name, start with how to get reviews on your website before adding a third OTA.

Doing this with JustReview

We import this source ourselves, which is why the numbers above exist at all, and it is the reason to pick a tool that runs its own connector instead of reselling a feature list. Practically, that means the half-star ratings are rendered as they actually are rather than rounded up, the language filter is available per widget so a German page shows German reviews, and the liked and disliked halves are visible to you before anything goes public, so the complaint inside a five-star review is a decision you make rather than a surprise you discover. Your full history arrives, not the most recent page. The Booking.com reviews widget page has a live demo running on real data, and you can create an account and connect your property without a card.

Two honest limits, because you will hit both. A property with several years of history imports as a job that runs rather than finishing while you watch. And Booking is genuinely slow: if you expect your widget to change weekly, it will not, so pair it with a source that moves. Plans and the review limits attached to them are on the pricing page.

Your guests already wrote the case for your hotel. It is sitting on a page that charges you commission every time someone reads it, and moving it to the page where the booking is free is the cheapest work on your list this week.