Occupancy rate is the share of your available nights that actually booked over a period. You calculate it by dividing booked nights by available nights. If a unit was available 30 nights last month and booked 18 of them, its occupancy rate was 60%. It is the most intuitive metric in short-term rentals, the easiest to see, and for exactly those reasons the most dangerous one to manage to.
Every owner understands occupancy instinctively. A full calendar feels like success and an empty one feels like failure. That instinct is the problem, because occupancy says nothing about price. A calendar can be full because demand is strong, or full because you priced it too low. The number looks identical either way.
"A full calendar feels like winning. Sometimes it just means you sold every night too cheap."
So what is a good occupancy rate?
The honest answer is that there is no single good number, and the right target depends on your market, your season, and your rate. As a rough orientation, many leisure short-term rental markets run a healthy annual occupancy somewhere in the 50 to 70% range, with peak months far higher and shoulder months far lower. Urban markets often run higher and steadier. Highly seasonal lake, beach, and ski markets run lower on an annual basis because the off-season is genuinely quiet, and that is fine.
But the annual average is the least useful version of the number. The real question is never what is a good occupancy rate in the abstract. It is whether your occupancy is right for the rate you are charging, on the dates in question, against your comp set. That is a different question, and it has a real answer.
Why 95% occupancy is usually a red flag
Operators are often proudest of the units running 90 to 95% occupied. In most markets, that is the clearest sign of underpricing there is. If you are selling nearly every available night, the market is telling you it would have paid more for a meaningful share of them. You captured volume and gave up rate, and because the calendar is full, it never registers as a loss.
A property running 95% occupied at a rate $80 below its comp set is not winning. It is leaving money on the table on almost every night and disguising it as a packed calendar. The owner feels great and earns less than they should. The fix is to raise rate until occupancy settles into a band where you are capturing both rate and fill, which in most leisure markets lands well below 95%.
"In most markets, 95% occupancy is not a trophy. It is the market telling you it would have paid more."
The three numbers that make occupancy meaningful
Occupancy on its own is an input, not a result. To know whether yours is good, read it next to three other things.
ADR.
Occupancy and average daily rate trade against each other. High occupancy at a low rate and low occupancy at a high rate can produce the same revenue, or wildly different revenue. You cannot judge one without the other in view.
RevPAR.
Revenue per available night, ADR multiplied by occupancy, is the number that settles the tradeoff. It rewards the combination of rate and fill that actually earns the most, and it cannot be gamed by chasing either one alone. This is the metric to manage to.
Pace against the comp set.
How your forward occupancy compares to the market's forward occupancy for the same future dates, right now. Pace tells you whether an open calendar is a real problem or just normal early-booking behavior, weeks before occupancy resolves.
What do I do if my occupancy looks low?
First, do not reflexively cut price. Low occupancy is a symptom, and price is only one of its causes. Work the diagnosis in order, the same way we do on a managed book.
- 01Check pace, not just occupancy. A calendar that looks empty 40 days out may simply be early. Read your forward position against the market before you conclude anything is wrong.
- 02Check structure. A minimum-stay rule that is too long, or length-of-stay settings that orphan nights, will block bookings no price cut can recover. More soft weeks are caused by a bad min-stay than by a high rate.
- 03Check distribution. Confirm the unit is visible on every channel it should be, at the right rate, ranking in search. A listing that fell out of search is a distribution leak, not a pricing problem.
- 04Only then consider a targeted rate move, on the specific soft dates, in the size the gap calls for. Never a blanket discount across a calendar that was mostly going to fill.
This ordering is the whole difference between revenue management and panic discounting. A 40-unit condo operator we manage on Banderas Bay in Puerto Vallarta came to us at 32% occupancy. Seventeen months later the same units were running 51%, with the nightly rate held flat the entire time, and same-store Adj. RevPAR had climbed from $51 to $81, a 58% gain on the KeyData same-store methodology. None of that started with a price cut. It was pace, structure, and distribution, worked in exactly this order.
Frequently asked questions
What is a good occupancy rate for an Airbnb?
Many leisure short-term rental markets run a healthy annual occupancy in the 50 to 70% range, with peak months far higher and shoulder months far lower. Urban markets run higher and steadier; highly seasonal lake, beach, and ski markets run lower annually, and that is normal. The real test is whether your occupancy is right for the rate you charge, on the dates in question, against your comp set.
Is 95% occupancy good?
Usually it is a warning sign. Selling nearly every available night means the market would have paid more for a meaningful share of them. In most leisure markets the profitable band, where you capture both rate and fill, sits well below 95%.
How do I calculate my occupancy rate?
Divide booked nights by available nights over the period. A unit available 30 nights that booked 18 of them ran 60% occupancy. Read it next to ADR and RevPAR before judging it.
So the question worth sitting with is not whether your occupancy is high. It is what your RevPAR would have been if every one of those booked nights had been priced right. We can pull that answer from your own data, benchmarked against your comp set, same-store, for free.
Adapted from Pacer's editorial archive, January 2026.