How a hotel's Booking.com review score works

How Booking.com calculates a hotel's review score

This is what Booking.com says on its official guest reviews page, checked on 23 October 2026:

Booking.com does not publish how much each review counts by age. Some guides say the score is the average of six categories (cleanliness, comfort, location, facilities, staff and value for money); according to Booking.com's page, the overall score and the subscores are given separately, with no direct link between them.

What this means for your hotel

• What happens today counts more than what happened two years ago. A problem you fix this month shows up sooner than it would if every review counted the same. So does a new one.
• Subscores do not drive the score, but they tell you where to look. If cleanliness or value for money sits below the rest, there are complaints there to read.
• Only Booking.com guests count. What people say on Google or TripAdvisor does not feed this score, although the complaints are usually the same. The Google and TripAdvisor guide explains how those two work.
• The score cannot be bought. Offering anything in exchange for a review breaks Booking.com's rules, and fake reviews are deleted.

Illustrative example. A hotel has 300 reviews averaging 8.0. The table shows where the average would land after the next 60 reviews if every review counted the same. Because Booking.com gives recent reviews more weight, the real change would be somewhat larger in either direction, but Booking.com does not say by how much.

Why it matters: a study by Chris Anderson for Cornell's Center for Hospitality Research (The Impact of Social Media on Lodging Performance, 2012), using Travelocity booking data, estimated that a hotel that raises its review score by one point on a 5-point scale can raise its price by 11.2% and keep the same occupancy. It is a 2012 figure from another platform; treat it as an order of magnitude, not a forecast for your hotel.

How to get better reviews on Booking.com

Nobody can promise how much your score will rise or when. What you can control is what you fix and whether it works:

1. Read recent reviews, not three years of them. Take the last 90 days on Booking.com, and ideally Google and TripAdvisor too, and log each complaint with its topic: breakfast, noise, hot water, front desk manner. Low subscores tell you where to start.
2. Count how many reviews repeat each complaint. A complaint in 40 reviews matters more than one in 3, however harsh those 3 are. Rank the list: at the top, what comes up a lot and is cheap and quick to fix; renovations come later. The guide on what to fix first explains how to rank it.
3. Turn each complaint into a specific change with an owner. "Improve breakfast" is not a change; "the head chef tastes every tray before the buffet opens" is. Give each one a person and a way to check it is done (a signed sheet, a check every morning).
4. Make the listing match what guests find. If complaints say "not what I expected" (smaller room, no lift, paid parking, a noisy street), the fix is in the photos and description on the listing, not in the hotel.
5. Catch problems before the guest leaves. Ask after the first night whether everything is fine, and keep an issue log at the front desk with the time and an owner. A shower fixed on day one does not turn into a 6 out of 10.
6. Reply to reviews, especially negative ones, with what you changed. A reply does not change that guest's rating, but future guests read it. Proserpio and Zervas (Marketing Science, 2017) found, using TripAdvisor reviews, that hotels that started replying gained 0.12 stars on average and received 12% more reviews. The guide on how to reply to hotel reviews has examples.
7. Check whether it worked. Four and eight weeks after a change, see whether that complaint appears less in new reviews. If it does not drop, that was not the fix, or it is not being done.

What to avoid

• Offering anything in exchange for a review, whether a discount, a drink or a gift. Booking.com prohibits it for properties.
• Writing reviews on behalf of guests or asking friends for them. Booking.com checks that reviews are genuine and deletes fake ones.
• Chasing the category average. Guests give the overall score separately; subscores help you find complaints, not calculate the score.
• Spending on the expensive things first. A renovation takes months to show up in reviews; a cheap change that solves a repeated complaint shows up sooner, and recent reviews are the ones that count most.

An example from real reviews

This is the result of a PolarPuffin test report for an all-inclusive beach hotel, with 204 reviews with text from Google, TripAdvisor and Booking.com over the last 90 days. The hotel is not named. Each change comes from complaints repeated in the reviews, and what to expect is an estimate of the 90-day Booking.com score, never a promise:

The last two changes will not move the score visibly, and the report says so. They are there because they remove complaints that keep coming back and cost little to do.

Where PolarPuffin fits

PolarPuffin is a hotel review management tool that does steps 1, 2, 3 and 7 of this guide every week: it reads Google, TripAdvisor and Booking.com reviews, counts which complaints repeat, suggests up to five small, cheap changes, each with an owner and the reviews behind it, and checks at 4 and 8 weeks whether those complaints went down. With Hotel Pro it also writes draft replies that only state what the hotel has confirmed.

PolarPuffin cannot raise your score for you, does not ask guests for reviews and does not change your Booking.com listing: the hotel does that. If you want to reply to a high volume of reviews or send surveys, the review management software page compares the tools that do.

PolarPuffin costs $79 (Hotel) or $149 (Hotel Pro) a month per hotel, excluding VAT, with no minimum term, and you can try it for 30 days with no card, using the hotel's real reviews. More on the how it works and pricing pages.