Hotel metrics: occupancy, ADR, RevPAR, TRevPAR and comp sets

Hotel brokers speak in a handful of metrics, and a buyer who knows exactly what each one counts is hard to mislead. Occupancy, ADR and RevPAR describe how the hotel sells rooms; TRevPAR describes the whole business; competitive-set indices describe how it performs against neighbours. This article defines them, shows how to read seasonality and market supply, and lists the checks that expose weak numbers.

The core definitions

• Occupancy: rooms sold divided by rooms available, for a period. Rooms given free, used by staff or held out of order need clear treatment: complimentary rooms are usually counted as sold at zero revenue, rooms out of order are normally still counted as available unless the method states otherwise.
• ADR, average daily rate: room revenue divided by rooms sold. Check whether revenue is net of VAT and of commissions, and whether breakfast is included. An ADR that is gross of tax or of channel commission is not comparable with one that is net.
• RevPAR, revenue per available room: occupancy multiplied by ADR, or room revenue divided by rooms available. It rewards neither empty rooms nor discount-driven volume.
• TRevPAR, total revenue per available room: total hotel revenue (rooms, food and beverage, spa, parking and other) divided by rooms available. It matters where non-room income is large: resorts, conference hotels and city hotels with restaurants. Two hotels with the same RevPAR can have very different TRevPAR and very different profit; see how to read a hotel P&L.
• GOPPAR, gross operating profit per available room: the profit counterpart, useful for comparing operators.

None of these says anything about cost. A rising RevPAR achieved through higher commissions or heavy marketing spend can leave profit flat.

Seasonality: annual averages hide the business

Annual occupancy of 65 percent can describe a steady city hotel with 60 to 70 percent every month, or a resort that is 95 percent full for four months and nearly empty for five. They carry very different risks: the resort needs seasonal staff, working capital for the winter and weather luck. Always ask for monthly occupancy, ADR and RevPAR for three to five years and for the day-of-week pattern. Check which months produce most of the profit, how fast the hotel fills in the shoulder months and whether a single event, such as a trade fair, a festival or a conference, drives a large share of demand. Strong event dependence is a risk if the event moves, shrinks or is cancelled.

The competitive set and STR/CoStar benchmarking

A competitive set (comp set) is a defined group of hotels that compete for the same guests, usually four to six properties of similar location, class and size. STR, now part of CoStar, collects data from participating hotels and returns anonymised, aggregated results, so a hotel sees how it performs against its set without seeing individual competitors. Three indices summarise it, each calculated as the hotel's figure divided by the comp-set average, multiplied by 100.

• MPI, market penetration index: the occupancy index.
• ARI, average rate index: the ADR index.
• RGI, revenue generation index: the RevPAR index, also known as the fair-share index. It is approximately MPI multiplied by ARI divided by 100.

An index of 100 means a fair share; above 100 the hotel outperforms its set, below 100 it underperforms. The combination matters: an MPI of 110 with an ARI of 85 says the hotel fills rooms by discounting. An MPI of 85 with an ARI of 115 says it holds price but loses volume.

A hypothetical illustration, with round numbers: a hotel runs at 66 percent occupancy and an ADR of EUR 110, while its comp set averages 70 percent and EUR 120. MPI is 66 divided by 70, about 94. ARI is 110 divided by 120, about 92. RevPAR is EUR 72.60 against EUR 84.00 for the set, so RGI is about 86. The hotel gets only 86 percent of its fair share. A seller will say the upside is in closing that gap; a buyer should ask why it exists (condition, reviews, location, distribution or management) and what it would cost to close.

The quality of an index depends entirely on the set. A seller who picks weaker neighbours or leaves out the strongest competitor flatters the hotel. Ask for the list of hotels in the set, check that they are comparable in class and location and, if possible, build your own. Smaller markets may have no STR coverage at all; in that case ask for the hotel's own booking data and for municipal or tourism-board statistics.

Market data and demand drivers

Hotel performance sits inside a market, so read the market first. Eurostat publishes nights spent in tourist accommodation by country and region (Eurostat tourism statistics), a starting point for demand trends, seasonality and the share of foreign guests. Broker and consultancy reports such as those of HVS, CBRE, JLL and Cushman & Wakefield, and STR/CoStar monthly data, add RevPAR trends by city. As an anchor, reports based on STR data indicated European RevPAR growth of about 3 percent in the first half of 2026, with ADR up about 2.2 percent, while forecasts for the year were more modest; the same reports note that inflation and financing costs may limit further gains. Use such figures as context only; your hotel's own city, segment and season matter more than a continental average.

Ask what drives demand in the specific location.

• Leisure: climate, beaches, attractions, flight connections and the air-route network, visa regimes and the image of the destination.
• Corporate and group: companies, universities, hospitals, trade-fair centres and conference venues.
• Events and infrastructure: a new airport terminal, a metro line or a convention centre can lift demand; closure of a large employer or route can remove it.
• Source markets: the nationality mix of guests and its exposure to currencies, travel advisories and airline capacity.

Supply pipeline

New hotels reduce the performance of existing ones, especially in similar classes and in small markets. Check the pipeline: rooms under construction and planned within the next three years, as a share of existing stock, by class. Municipal planning offices, STR pipeline reports and local press give figures. Also watch the shadow supply: short-term rentals, which often grow faster than hotels in leisure destinations, and conversions of offices and residential buildings into hotels. A market with rising demand and a large pipeline can still see falling RevPAR.

How to sanity-check a seller's numbers

• Start with three years of monthly actuals, not a forecast or the best year. Ask for the budget and compare it with what happened.
• Reconcile revenue with bank statements, VAT returns and the property management system; differences need explanations.
• Compare occupancy and ADR with the comp set and with local statistics. A hotel that claims 85 percent occupancy where the market averages 60 percent needs evidence.
• Check ADR against what the same rooms are offered for online on typical dates, and look at the channel mix and the commission share. See the due diligence checklist.
• Look at the effect of one-offs: groups, events, a renovation that closed rooms, a pandemic-era comparison year.
• Stress the case: occupancy down 5 points, ADR flat or down a few percent, costs up with inflation. If the hotel only works in the best case, the price is too high; see how to value a hotel for sale.

FAQ

Is RevPAR enough to compare two hotels? No. It ignores non-room revenue and all costs. Add TRevPAR, GOP margin and GOPPAR before drawing conclusions.

What is a good RGI? Above 100 means a hotel gets more than its fair share. The size of the gap matters less than why it exists and whether it is sustainable.

Can a small hotel get STR data? Participation is generally for hotels that report their data, and coverage varies by country. Ask the seller whether the hotel subscribes, and use local statistics if it does not.

How many years of data should I ask for? At least three years of monthly figures, plus year-to-date and forward bookings.

Do the metrics differ by country? Definitions are broadly standard, but treatment of taxes, service charges and breakfast differs. Ask how the numbers are calculated before comparing.

How we help

We are a trilingual platform that finds properties and coordinates independent local lawyers, accountants and tax advisers for buyers who read Russian, Polish and English. We do not provide market research subscriptions or guarantee performance. This text is general information, not legal, tax or investment advice; market data changes, so verify current figures with the named sources and your advisers.