RevPAR Explained: Formula, Benchmarks, and the Cost It Doesn't Show

RevPAR stands for revenue per available room. It is the standard measure of how well a hotel turns its room inventory into money, and it is the figure most market benchmarking reports are built around.

The first formula: RevPAR equals total room revenue divided by the number of room nights available in the period. A room night available is one room for one night, so a 40-room hotel has 40 available room nights per day, or 1,200 in a 30-day month.

The second formula: RevPAR equals average daily rate (ADR) multiplied by occupancy. ADR is the average revenue per occupied room, and occupancy is the share of available rooms sold. The two formulas are the same calculation arranged differently, so they always give the same result if the inputs are consistent.

Two details trip people up. Use room revenue only, not food, beverage or spa revenue, and decide whether you are reporting it with or without VAT and keep that fixed. And use rooms available, not rooms sold: a room you took out of service for renovation is usually excluded, but a room you simply failed to sell must stay in the denominator, because that is the point of the measure.

An independent hotel with 40 rooms, over a 30-day month. These figures are illustrative, chosen to be plausible for a mid-range European city hotel.

Rooms available: 40 rooms × 30 nights = 1,200 room nights. Rooms sold: 900, so occupancy is 900 ÷ 1,200 = 75%. Room revenue: €126,000, so ADR is €126,000 ÷ 900 = €140.

RevPAR by the first formula: €126,000 ÷ 1,200 = €105. By the second: €140 × 75% = €105. Same answer.

What the number tells you: across every room you had to sell, you earned €105 per room per night on average, counting the empty ones as zero. That is why RevPAR is more useful than ADR or occupancy alone. It punishes both empty rooms and cheap ones.

Occupancy on its own rewards filling rooms at any price. A hotel can push occupancy to 95% by discounting heavily and end up with less revenue than at 75%. If a revenue manager reports rising occupancy without the rate, ask for the rate.

ADR on its own rewards high prices regardless of how many rooms sell. A hotel can hold a high ADR by refusing to discount and leave half the building empty. ADR is meaningful only next to occupancy.

RevPAR combines the two, which is why it became the default. But it has its own blind spots. It ignores everything outside rooms, so a hotel with a strong restaurant or spa looks the same as one without. It ignores costs entirely, including what it cost to sell each room. And it does not care how the booking arrived, so a room sold through an OTA at €140 counts exactly the same as a room sold on your own website at €140, even though you keep noticeably less of the first one.

That last blind spot is the one this guide is about, because it is the one that hides a strategy problem behind a healthy-looking number.

Net RevPAR, sometimes written NRevPAR or NetRevPAR, is room revenue minus distribution and acquisition costs, divided by rooms available. Distribution costs include OTA commission, booking engine and payment fees, and travel agent commission. Some hotels also subtract channel-specific marketing such as metasearch clicks.

The idea has been pushed for years by industry analysts who argued that gross RevPAR overstates performance when acquisition costs rise. Cornell's School of Hotel Administration has written about Cindy Estis Green of Kalibri Labs and Mark Lomanno advocating net ADR and net RevPAR, measured after all commissions, for exactly that reason.

How big the gap can be: Kalibri Labs' study of US hotels, published by the American Hotel & Lodging Association, found customer acquisition costs had risen from 5% to 10% of guest-paid revenue in the 1990s to 15% to 25% in the US by 2016, with an average around 16% to 18% and some hotels at 35% to 40%. Those are US figures from a decade ago and include chain hotels' internal sales costs, so do not apply them to your hotel. They show why a gross number can drift away from what a hotel actually keeps.

You do not need an industry study to calculate your own. Take a month of room revenue by channel from your PMS, subtract what each channel cost you, and divide by rooms available.

Three 40-room hotels, each with the same month as the worked example: 1,200 rooms available, €126,000 room revenue, RevPAR €105. The only difference is how the rooms were sold. We assume OTA bookings cost 18% in commission and direct bookings cost 5% in booking engine and payment fees, the same assumptions Staylight's calculators use. Neither is a published industry rate. Use your own.

Hotel   OTA share   OTA cost   Direct cost   Total cost   Net revenue   RevPAR   Net RevPAR
A       85%         €19,278    €945          €20,223      €105,777      €105     €88.15
B       60%         €13,608    €2,520        €16,128      €109,872      €105     €91.56
C       30%         €6,804     €4,410        €11,214      €114,786      €105     €95.66

On a RevPAR report these three hotels are identical. On net RevPAR, hotel C keeps €7.51 more per available room per night than hotel A. Between B and C the gap is €4.10, which across 1,200 room nights is €4,914 for the month and about €59,000 over a year with the same pattern.

Nothing about the rooms, the rates or the occupancy differs. The difference is entirely where the guests booked. That is the strongest argument for reporting net RevPAR next to RevPAR every month: it is the only standard number that moves when you shift channel mix, and channel mix is the lever most within a hotel's own control.

Channel mix is decided before the guest reaches you. If the first place a guest sees your hotel is an OTA listing, there is a good chance the booking goes through the OTA. If the first place is your own site, or a search result that links to it, the odds shift.

A growing share of travellers now start by asking an AI assistant where to stay. When ChatGPT, Gemini or Google's AI Mode names a hotel, the answer often links to an OTA rather than the hotel's own website, because an OTA page is a dense, frequently updated source engines trust. Our guide on why AI recommends Booking.com instead of your hotel explains the mechanism.

So AI visibility is a channel-mix question. A hotel that is named in AI answers and cited from its own site has a chance to win those guests direct. A hotel that is absent, or named only through an OTA link, hands that share to the OTA by default. None of that shows up in RevPAR. It shows up in net RevPAR.

If you want to put a number on it, our revenue-at-stake calculator estimates how much direct revenue an AI visibility gap may be costing, using your own bookings, and our OTA commission calculator totals what you currently pay in commission.