Occupancy is the easiest hotel metric to improve and the easiest one to improve for the wrong reasons. Drop your rates far enough and you will fill every room, which is why a full property is not, on its own, evidence of anything.
Here is the formula, and the four situations where the number lies to you.
The calculation
Occupancy = rooms sold / rooms available
Both sides are room nights, not rooms. A 20 room property over a 30 night month has 600 available room nights. Sell 390 of them and you are at 65%.
390 / 600 = 65%
Two details decide whether your number is comparable to anyone else's. Count a room as available even when it is out of order, unless you are reporting internally and have said so, because excluding maintenance rooms quietly inflates the figure. And count the room, not the beds: a hostel reporting bed occupancy and a hotel reporting room occupancy are not measuring the same thing.
Where it misleads
It says nothing about what you charged. This is the big one. Occupancy and ADR move against each other, and the only figure that holds both is RevPAR, which is room revenue divided by rooms available. At 65% and €110 you are at €71.50 RevPAR. Fill to 85% by cutting to €80 and you land at €68, so you sold 120 more room nights, did more laundry, served more breakfasts, and earned less. The occupancy chart went up and to the right the whole time.
100% means you were too cheap, at least on the night it happened. A sold-out night is a night where demand exceeded your price and you never found out by how much. One sell-out in a season is good luck. A pattern of them is a pricing problem, and the fix is to find out what the market was charging for those exact dates.
A monthly figure hides the shape of the month. 65% occupancy can be a steady 65% every night, or three full weekends and empty midweeks. The second needs midweek offers and tighter weekend pricing; the first needs something else entirely. We tracked how much individual nights move in the summer 2026 rate report, and the short answer is: a lot more than the monthly averages do.
Booking pace matters more than the final figure. Being at 40% for next Saturday is fine in January and a problem on Thursday. Occupancy without a lead time attached is a result, not a signal. What turns it into a signal is knowing whether comparable properties are also at 40%, or whether they filled last week while you were waiting.
What to check alongside it
For any date where occupancy is telling you something, check three things before you change a price.
Competitor rates for that date. Not their average, that specific night. If you are soft and they are too, it is the market; if you are soft and they are full, it is you.
Whether they still have rooms. This is the one that separates a filling market from a mispriced rival. Competitors with no availability left and rates climbing means you can hold or raise. Competitors with everything open at a price below yours is a different diagnosis entirely, and we wrote up the reasoning in availability monitoring.
Minimum stay restrictions. A two night minimum on a Saturday removes single night demand from the market. If rivals have one on and you do not, your occupancy will look better than theirs for reasons that have nothing to do with price.
Picking an occupancy target
The honest answer is that there is no universal target, because the right one depends on your cost per occupied room and what you give up to reach it. The useful version of the question is narrower: at what occupancy does my RevPAR stop improving?
Work it out from your own numbers. Take last season's nights, group them by occupancy band, and compare RevPAR per band. Most independent properties find a plateau somewhere in the 70s or 80s, past which the extra room nights arrive only at rates that pull the average down. That plateau is your target, and it is yours, not the one in an industry benchmark report. Keep a floor and a rack rate ceiling around it so a rule-driven change cannot run away.
Getting the comparison data
The hard part of this is not the lookup, it is the comp set. Five properties that a guest would genuinely consider instead of yours, in your band, at your distance from whatever people come for, is a short list and it takes judgement to build. For rentals that judgement is the whole subject of Airbnb competitor analysis, where the usual mistake is comparing a two bedroom apartment to a villa because both are nearby.
Once the list exists, OTABot keeps it current: rates, availability and minimum stay for each check-in date, read from the public Booking.com and Airbnb listings, with alerts when one of your five moves. The hotel competitor pricing checklist is the weekly routine to run against it, and Booking.com price tracker and Airbnb rate monitor cover each platform.
The short version
Occupancy is rooms sold over rooms available, and on its own it is a volume metric with no price in it. Read it next to RevPAR, read it per date rather than per month, and read it next to what comparable properties were charging and whether they had rooms left.
Track up to two competitors free on OTABot.