How hotels are rated

There is no single, worldwide hotel rating. The stars on a booking site, the stars on the door, a Forbes rating and a Michelin Key are four different things, issued by four different kinds of organisation, measuring four different things. That is why the same hotel can be 4-star on one site and 5-star on another, and why both can be honest.

This guide sorts them into the three families that exist, says what each one measures, and says what each one cannot tell you.

1. Official star classification: what the hotel has

In much of Europe, stars are an official classification. A national body or a hotel association checks the property against a published list and awards one to five stars.

The largest shared scheme is the Hotelstars Union, which covers 21 European countries with one catalogue of 239 criteria, in force since January 2025. Some criteria are mandatory for a given star level and the rest earn points; a hotel needs the mandatory items and a minimum score. France and Portugal run national systems of their own. Finland and Norway have no star classification at all. The United States has no government star classification either: its ratings come from private bodies and from booking sites.

What it measures: facilities and service formalities. A lift, reception hours, how often linen is changed.

What it cannot tell you: how good the hotel is. Two hotels with the same inventory get the same stars, whether the stay is forgettable or superb.

2. Inspection ratings: how the stay performs

A second family sends trained inspectors who stay as ordinary guests and score what happens.

Forbes Travel Guide rates luxury properties as Five-Star, Four-Star or Recommended. Its inspectors are anonymous and test up to 900 standards, weighted toward service.

AAA Diamonds cover North America. Inspections are unannounced, and a property is designated Approved, Three, Four or Five Diamond.

Michelin Keys, introduced in 2024, award one, two or three Keys on five criteria: architecture and interior design, service, character, value for the price, and contribution to the neighbourhood or setting.

What it measures: the delivered experience, by someone paid to notice.

What it cannot tell you: anything about the hotels nobody inspected. These programmes cover a few thousand properties.

3. Platform stars and guest scores: what the site decided

The stars on a booking site are usually the site's own category, sometimes taken from the official classification and sometimes declared by the hotel. Guest scores are averages of reviews on that platform, on that platform's scale, from that platform's customers.

What it measures: popularity among the people who booked there.

What it cannot tell you: how one platform's 8.7 compares with another's 4.4 out of 5, or how much a score built on a dozen reviews is worth.

Why the same hotel gets different ratings

Because the question changes. One system asks what the building contains, another asks how an inspector was treated, a third asks what past guests typed. A hotel can pass one and fail another without anyone being wrong.

Where HOTELSRATING fits

The HOTELSRATING Index (HRI) is a fourth kind of thing: a formula. It does not inspect hotels and it holds no guest reviews. It combines independent expert recognition (Forbes Travel Guide, Michelin Keys, The World's 50 Best Hotels, World Travel Awards) with a hotel's confirmed facilities into one score on a published scale of 20 to 99.5, recomputed nightly.

A hotel receives a score only when at least one independent body has recognised it. Every other hotel is published with its facts and sources, and no score. The formula is public and deposited under a permanent identifier, so anyone can check how a score was produced. Nothing a hotel can buy from us changes its score, medal or position.

Read the methodology. Read the Independence Charter.

Questions people ask