How to calculate hotel occupancy rate (and the trap of looking at it alone)

By Arshad Kacchi · Founder, RevPerfect · 16 May 2026 · 8 min read

The first time a GM walked me through the morning numbers, he pointed at the occupancy line and said "that's the only one that matters." We were sitting at a 92% house, and he looked pleased with himself. Six months later I learned how to calculate hotel occupancy rate properly — and within a year I'd learned the harder lesson: an occupancy line on its own is the easiest number in the building to misread. A hotel can run at 92% and still be the worst performer in its comp set. A hotel can run at 68% and be the most profitable property in the market. The arithmetic is trivial. The interpretation is where the money is.

What hotel occupancy rate actually means in 2026

Occupancy is the share of your inventory that was sold. Take the rooms you had available, count the ones occupied by a paying guest, divide one by the other, multiply by a hundred. That is the whole concept.

It is the oldest performance metric in hospitality, predating ADR, RevPAR and GOPPAR. It survives because it answers a question every owner instinctively asks: was the hotel busy? Owners understand it. General managers report on it. Banks ask for it in covenants. It is one of the three figures every hospitality dashboard in the world is built to display.

It is also, when read on its own, the most dangerous metric in the stack.

The hotel occupancy rate formula

Here is the working version. There is only one.

Two inputs, both of which need a definition.

Occupied rooms means rooms sold with a paying guest in them for the night. House-use rooms — held for staff, maintenance crews, or owner-personal use — are usually excluded. Complimentary rooms typically stay in the numerator because they are occupied, even if no revenue flowed. The rule varies by hotel; write it down and apply it consistently.

Available rooms means total inventory for the period. For a single night, it is the room count of the property. For a week, it is the room count multiplied by seven. From that you subtract rooms genuinely out of order — rooms a maintenance ticket has taken offline. Out-of-service rooms (renovation staging, deep clean) are a judgement call; most operators keep them in the denominator because the decision was operational, not structural.

A worked example: 120-room property, one night

Run the calculation on a Saturday night at a 120-room property.

Occupancy is 98 ÷ 116, which is 84.5%. If your PMS shows a different number, walk back through the five inputs — nine times out of ten the disagreement is definitional, not arithmetic.

Across a full week, the arithmetic is unchanged but the denominator stretches: 116 rooms × 7 nights = 812 available room nights. If 612 were occupied, weekly occupancy is 75.4%. That single percentage hides the variance underneath — a 95% Saturday, a 56% Tuesday, a 38% Sunday — which is why I'd never read it without the day-of-week shape behind it.

Where occupancy breaks down (the trap)

Here is the part nobody puts in the textbook. Occupancy is a volume metric. It measures how much of the building was sold. It says nothing about what you sold it for. Two hotels can run at identical 85% occupancy — same week, same city, same comp set — and one can be earning thirty percent more revenue per available room than the other. The reason is rate, and rate is invisible inside an occupancy number.

Two 120-room hotels, same week.

Same occupancy. Forty-seven dollars per available room per night in the difference. Across a year on a 120-room property, that gap is roughly $2.06 million in rooms revenue Hotel A handed away by holding rate too soft for the demand it was getting. The occupancy line reports neither side of that — only that both buildings filled to the same level.

This is why I will not let a revenue manager I work with report occupancy without ADR and RevPAR sitting next to it. The three together are the minimum honest read.

An occupancy number on its own is a story with the middle chapter torn out. The hotel was busy. Whether it earned what it should have is a different question entirely.

I covered the interaction between rate, occupancy and revenue in detail in how to calculate RevPAR, which is the natural companion read. The short version: RevPAR equals occupancy multiplied by ADR. The moment you have all three on one line, the trap of looking at occupancy alone disappears.

The other thing occupancy hides: channel cost

The second failure mode is subtler. A hotel can lift occupancy by widening distribution — more OTA channels, accelerator programmes, wholesale contracts at a contracted markdown. The line will climb. The owner will be pleased. The cash position will not move the way the chart suggests.

The rooms added came in at higher distribution cost than the ones already there, and occupancy is blind to distribution cost. A high-occupancy, high-commission hotel can underperform a moderate-occupancy, low-commission hotel on every measure the owner cares about — net rooms revenue, contribution margin, profit per available room.

I covered the structural shape of channel cost in OTA commission rates in 2026. Short version: high occupancy bought through expensive channels is the most common form of margin leakage on independent properties. Invisible on the occupancy line. Brutal on the P&L.

What to do about it — a five-step playbook

The workflow I run with every property. Not complicated. Just rarely done.

1. Calculate occupancy daily, weekly and monthly using the formula above. Single-period occupancy is almost useless on its own — the trend is the signal, not the level.
2. Display occupancy, ADR and RevPAR on the same line on every dashboard, every owner pack, every morning report.
3. Show the day-of-week shape behind every weekly occupancy figure. A 75% week of seven 75% days is a different property than a 75% week with a 95% Saturday and a 38% Sunday.
4. Track the gap between gross and net. If occupancy climbs but net revenue per available room is flat, the lift came from more expensive channels.
5. Compare against same-week-last-year and same-week-last-month. Single-period occupancy without a reference frame is impossible to interpret.

Steps three and four are the ones most operators skip, and they are the ones that change how owners think about the property.

A real scenario — anonymised, but it happens every quarter

I worked with a 90-key independent in a coastal Australian market last year. The property had run hot for two summers — occupancy in the high eighties across peak season, the GM reporting full-house weekends, the owner pleased with the chart. On the surface, performing.

The owner asked me to look at the rate side because something felt off. We pulled the day-of-week shape behind the headline occupancy. Friday and Saturday were running 96% and 99% across the season. Sunday through Thursday sat in the high seventies. The weekend numbers were full to a point where they could not physically grow — the property was sold out, and the demand being turned away suggested rate had been left soft on the highest-yielding nights of the week.

We modelled what a $40 rate lift on weekend nights would have produced across the season. The maths landed at roughly $312,000 in additional rooms revenue — none of which would have shown up on the occupancy line, because the line was already at the ceiling. The headline metric was telling the owner the property was performing. The metric it actually needed — the rate it was leaving on the table on its high-demand nights — had been absent from the dashboard.

That property's revenue manager is competent. They had been doing the job to the level the standard dashboard demanded — and the standard dashboard did not include the question the property most needed answered. Forecasting demand precisely is the other half of this puzzle, covered in hotel demand forecasting for revenue managers if you want the companion read.

Is your dashboard showing you occupancy alone, or occupancy in context?

Walk over to whatever dashboard you use — your PMS report, your business intelligence layer, your weekly owner pack — and look at the occupancy figure. Then ask yourself two questions. Is ADR on the same line? Is RevPAR on the same line? If either is missing, you are reading half the story.

This is one of the gaps RevPerfect was built to close. Every operator surface we build shows occupancy alongside ADR and RevPAR, with the day-of-week shape underneath and the channel-mix context beside it. We don't tell you what to do with the answer. We just stop you from seeing the wrong number on its own.

For the strategic layer above the metric layer, the natural follow-on read is hotel revenue management strategies for 2026. Once you've got the occupancy-ADR-RevPAR trio anchored, the next step is the comparison across all three margin metrics in ADR vs RevPAR vs GOPPAR.

FAQ

The honest dashboard

The point of this article — and the point of RevPerfect — is that knowing how to calculate hotel occupancy rate is useful, not complete. The calculation is simple. The interpretation is where most operators slip. The metric tells you how full the building was. It does not tell you what you sold it for, what you paid to fill it, or what landed in the bank. The gap between those questions is where most of the margin in a modern hotel is won or lost.

If your dashboard isn't showing you ADR and RevPAR alongside occupancy, your dashboard is showing you a fragment of the story. We built RevPerfect to close that gap — to give operators a single surface that respects the headline and the truth behind it. Try RevPerfect free → or book a 20-minute walkthrough and I'll show you what your occupancy has actually been doing.