What is the difference between RevPAR, ADR, and occupancy?

ADR is the average price of a booked night. Occupancy is the share of your available nights that booked. RevPAR combines both into revenue per available night, and equals ADR multiplied by occupancy. ADR measures price, occupancy measures fill, and RevPAR measures the two together, which is why it is the hardest of the three to move with a single lever.

Which metric should a short-term rental operator optimize?

RevPAR, because it is the only one of the three that a single-lever move cannot fake: optimizing ADR alone rewards an empty calendar, and optimizing occupancy alone rewards discounting. But even RevPAR ignores cost, so the real target is RevPAR at an acceptable margin, not RevPAR at any cost. This is informational, not financial advice.

Can ADR go up while RevPAR goes down?

Yes, and it is common. If raising your nightly rate loses more occupancy than the higher price makes up for, RevPAR falls even as ADR rises. That divergence is the single clearest reason operators track RevPAR rather than watching the nightly rate alone, because the rate can look healthy while revenue per available night is shrinking.

Is RevPAR or occupancy more important for a vacation rental?

Neither in isolation. High occupancy at a rate below your cost floor loses money on every booking, and high RevPAR driven by heavy discounting can still shrink your margin. Occupancy is an input, RevPAR is the combined revenue result, and both have to be read against your cost floor and return, not chased on their own.

RevPAR vs ADR vs Occupancy: Which Metric When

By Daniel Carrow (pen name) Published Jul 20, 2026 guide

Three numbers claim to tell you how your short-term rental is performing, and picking the wrong one for the decision in front of you leads to confident, expensive mistakes. Raise rates on the strength of a high ADR and you can bleed occupancy. Chase occupancy because it is low and you can discount away your margin. ADR, occupancy, and RevPAR each answer a different question, and the skill is knowing which question you are actually asking.

TL;DR: ADR is the average price of a booked night (rewards rate, hides vacancy). Occupancy is the share of available nights that booked (rewards fill, hides price). RevPAR is revenue per available night, ADR x occupancy (rewards rate and fill together, hides cost). Optimize RevPAR, because it is the only one a single lever cannot fake, but read it against your cost floor so revenue growth is not margin loss in disguise. Informational, not financial advice.

The three metrics, defined

Keep the question attached to each number and the confusion goes away.

ADR (average daily rate). Rental revenue divided by the nights you booked. It is the average price of a sold night and it asks: what are the nights I sell worth? It says nothing about how many nights sold.

Occupancy. Booked nights divided by available nights. It is the share of your calendar that sold and it asks: how much of my available supply am I capturing? It says nothing about the price of what sold.

RevPAR (revenue per available night). Rental revenue divided by available nights, equivalently ADR x occupancy. It spreads revenue across every available night, booked or empty, and it asks: what is each night of my calendar actually worth on average? It combines the other two into one figure (see the RevPAR pillar for the full breakdown).

What each measures, rewards, and hides

The disagreements between the three are not noise. They come from what each one deliberately leaves out.

Metric	What it measures	What it rewards	What it hides	Use it to
ADR	average price of a booked night	a higher nightly rate	vacancy: unbooked nights are invisible	judge pricing power on the nights you sell
Occupancy	share of available nights booked	a full calendar	price: how cheap the fill was	judge demand capture and calendar fill
RevPAR	revenue per available night (ADR x occupancy)	rate and fill moving together	cost: it sits entirely above expenses	judge combined top-line performance

Read the “hides” column top to bottom and the hierarchy is clear. ADR hides vacancy, so occupancy exists to expose it. Occupancy hides price, so RevPAR exists to fold both together. RevPAR hides cost, so your break-even and return metrics exist to expose that. No single number is complete; each covers the previous one’s blind spot.

When to use each

Setting or reviewing your nightly rate: ADR. When the question is specifically about price, what your sold nights command, ADR is the right lens, especially compared against your own history and your direct comp set. Just never read it without occupancy beside it.

Diagnosing a demand or calendar problem: occupancy. When nights are going empty and you want to know how much of your available supply you are converting, occupancy is the direct measure. Low occupancy is a prompt to ask why, not an instruction to discount.

Judging overall revenue performance: RevPAR. When the question is “is my pricing-and-fill strategy actually working,” RevPAR is the answer, because it is the only one of the three that punishes a rate hike that costs too much occupancy and a discount that buys too little. It is the number to trend over time and benchmark against your submarket.

Deciding whether that performance is a good business: neither, alone. All three sit above the cost line. Carry them into a decision alongside your break-even occupancy and cash-on-cash return, or you can optimize a revenue metric straight into a thinner margin.

Same property, three different stories

Take one listing over a month and watch the metric you pick change the verdict. Illustrative figures, not a projection for any real property.

Baseline. 30 available nights, 18 booked, 3,600 dollars of rental revenue on a consistent basis.

ADR: 3,600 / 18 = 200 dollars
Occupancy: 18 / 30 = 60 percent
RevPAR: 200 x 0.60 = 120 dollars

The rate hike. You raise the rate and ADR climbs to 240 dollars, but bookings fall to 13 nights (43 percent occupancy).

ADR: 240 dollars (up 20 percent, looks great)
Occupancy: 13 / 30 = 43 percent (down)
RevPAR: 13 x 240 / 30 = 104 dollars (down 13 percent)

An operator watching ADR sees a win. An operator watching RevPAR sees the truth: the higher rate did not pay for the nights it cost. ADR rose while RevPAR fell, which is exactly the divergence RevPAR exists to catch, and exactly why a dynamic pricing tool tuned to defend a high rate can quietly shrink revenue. That failure mode is covered in when dynamic pricing hurts.

The discount. Now instead you cut the rate to 170 dollars and occupancy jumps to 80 percent.

ADR: 170 dollars (down)
Occupancy: 80 percent (up, looks great)
RevPAR: 170 x 0.80 = 136 dollars (up 13 percent)

Here the discount worked on revenue: RevPAR rose because the occupancy gain more than paid for the lower rate. But this is the case where you must check the cost floor, because a RevPAR that rose on volume can still sit below break-even if the rate dropped too far.

One property, three months, and depending on which metric you watched you would have congratulated or fired yourself. The number is not the performance. The question is.

The trap: optimizing one metric in isolation

Every one of these metrics can be maxed out in a way that destroys the business.

Max ADR by pricing for the top 10 percent of demand: a beautiful rate on a near-empty calendar.
Max occupancy by underpricing relentlessly: a full calendar that loses money on most nights.
Max RevPAR by discounting into volume without watching cost: rising revenue per available night, falling margin.

The discipline is to optimize RevPAR subject to a constraint: never below the rate that clears your cost floor, and always with an eye on the return the property owes your capital. RevPAR is the right primary metric precisely because it resists two of the three single-lever games. It just cannot resist the third, cost, which is why it is a member of the toolkit and not the whole toolkit.

Where this fits

Use the three together, each for its own question:

Understand what RevPAR is and why it beats watching rate or occupancy alone: the RevPAR pillar.
Run the arithmetic, cleanly and consistently: how to calculate RevPAR and ADR.
Make sure the rate under your RevPAR clears its costs: break-even occupancy.
Watch the pricing-tool failure mode this comparison exposes: when dynamic pricing hurts.

Disclaimer: This article is informational only and is not financial, investment, or tax advice. Figures in the worked examples are illustrative and not a projection for any specific property. Verify current market data against primary sources before pricing or underwriting. Information current as of July 2026.

Frequently asked questions

What is the difference between RevPAR, ADR, and occupancy?: ADR is the average price of a booked night. Occupancy is the share of your available nights that booked. RevPAR combines both into revenue per available night, and equals ADR multiplied by occupancy. ADR measures price, occupancy measures fill, and RevPAR measures the two together, which is why it is the hardest of the three to move with a single lever.
Which metric should a short-term rental operator optimize?: RevPAR, because it is the only one of the three that a single-lever move cannot fake: optimizing ADR alone rewards an empty calendar, and optimizing occupancy alone rewards discounting. But even RevPAR ignores cost, so the real target is RevPAR at an acceptable margin, not RevPAR at any cost. This is informational, not financial advice.
Can ADR go up while RevPAR goes down?: Yes, and it is common. If raising your nightly rate loses more occupancy than the higher price makes up for, RevPAR falls even as ADR rises. That divergence is the single clearest reason operators track RevPAR rather than watching the nightly rate alone, because the rate can look healthy while revenue per available night is shrinking.
Is RevPAR or occupancy more important for a vacation rental?: Neither in isolation. High occupancy at a rate below your cost floor loses money on every booking, and high RevPAR driven by heavy discounting can still shrink your margin. Occupancy is an input, RevPAR is the combined revenue result, and both have to be read against your cost floor and return, not chased on their own.