RevPAN is revenue per available night: total rental revenue divided by every night a property was available to book, whether it booked or not. It is the single most useful revenue metric in short-term rental analysis because it collapses nightly rate and occupancy into one number, and it is Step 1 of the analysis sequence.
How is RevPAN calculated?
RevPAN equals average daily rate multiplied by occupancy. A property with a $310 ADR at 66 percent occupancy has a RevPAN of $204.60. Multiply by 365 to annualize, which produces roughly $74,700 of gross revenue. The alternative arithmetic reaches the same place: divide total annual revenue by 365. If the property was genuinely unavailable for part of the year, for renovation rather than owner use, divide by available nights instead.
Why is RevPAN more useful than ADR or occupancy alone?
Because either metric alone can be moved without improving income, and frequently is. Raising rates lifts ADR while thinning the calendar. Discounting fills the calendar while destroying rate. Both look like progress in isolation. RevPAN is the number that tells you whether either move actually produced revenue.
| Property | ADR | Occupancy | RevPAN | Annual gross |
|---|---|---|---|---|
| A | $600 | 32% | $192 | $70,080 |
| B | $240 | 68% | $163 | $59,568 |
| C | $310 | 66% | $205 | $74,679 |
Property A has nearly double the nightly rate of Property B and produces more revenue. Property C has roughly half the rate of Property A and produces the most revenue of the three. A buyer screening on ADR would rank these in exactly the wrong order.
How do you build a RevPAN estimate from comparable properties?
Select comps that match on the variables that actually drive booking in the specific submarket, not on the variables that are easiest to filter. Bedroom count and guest capacity are necessary but not sufficient. In most leisure markets the amenity set does more work than square footage: private pool, hot tub, ski-in access, waterfront, pet policy. Pull trailing twelve month performance for those comps, discard the top and bottom of the distribution rather than averaging everything, and take the median rather than the mean. Short-term rental revenue distributions are right-skewed, so the mean flatters the estimate.
Should you underwrite to the median comp or below it?
Below it, unless the subject property is already operating at or above median with documented history. A new owner inherits a listing with no review history, no calendar momentum, and no search ranking on the platform. First-year performance below the market median is the normal outcome, not the pessimistic one. Underwriting to the median assumes an operator who is already good at the job, running an asset that has already earned its position.
What RevPAN is high enough?
RevPAN alone does not answer that. Convert it to gross yield by annualizing and dividing by purchase price, which is what makes properties in different price bands comparable. A $74,700 gross on a $550,000 purchase is a 13.6 percent gross yield. Gross yield is a screening filter, not a decision. It tells you which properties survive to the expense stack, and the expense stack is where most of the apparent yield goes.
How does RevPAN differ from RevPAR in hotels?
They are the same construction applied to different units. Hotels measure revenue per available room per night across a portfolio of rooms in one building. Short-term rentals measure revenue per available night for a whole property. The distinction matters when comparing a short-term rental to hotel market data, because a hotel RevPAR of $150 and a whole-home RevPAN of $150 describe entirely different assets.
What comes after the revenue estimate?
Verification. An independently built RevPAN is only useful if it is then tested against what the seller claims, which is Step 2, and against the full expense stack in Step 3. Revenue is the input most often wrong in short-term rental underwriting, and it is wrong in one direction.
How does seasonality distort a RevPAN estimate?
An annual RevPAN figure hides the shape of the year, and the shape determines whether the property survives its own cash flow. A ski market may produce 70 percent of its revenue in four months. A Gulf coast property may produce most of its income between March and August. The annual average is correct arithmetic and a poor operating guide, because debt service, insurance, and utilities arrive in the off months at exactly the same rate they arrive in peak season. Build the estimate monthly, then annualize, and look at the trough. The relevant question is not what the property earns in a year but how many consecutive months it operates below breakeven and whether the reserve covers them.
What is the difference between gross RevPAN and net RevPAN?
Gross RevPAN uses total booking revenue before any deduction. Net RevPAN uses what actually reached the owner’s account after platform commission, and sometimes after management fees as well. Third-party data providers do not all define this the same way, and the difference is roughly 3 to 25 percent depending on where the line is drawn. Before importing any comp figure into a model, confirm which definition produced it. A gross RevPAN compared against a net expense stack overstates income by the entire commission load, and that single mismatch is enough to move a deal from failing to passing.
How many comparable properties do you need?
Enough that removing any single comp does not move the median materially, which in practice means eight to fifteen in most submarkets. Below roughly six comps the estimate is driven by whichever properties happened to be well managed that year. Above twenty the comp set has usually widened past genuine comparability and is picking up properties in different micro-locations, different amenity tiers, or different guest segments. If a submarket cannot produce six genuine comps, that is itself a finding: thin comp sets and thin resale markets tend to be the same markets.
How does RevPAN change in the first year of new ownership?
It usually falls, and buyers who model otherwise are modeling someone else’s business. A transferred listing may lose its review history depending on how the transfer is handled, and it always loses calendar momentum, repeat guests, and whatever search position the prior operator had built. Platform ranking rewards booking velocity and review volume, both of which reset in ways that take months to rebuild. A first-year discount of 10 to 20 percent against the stabilized comp median is a normal assumption, not a conservative one. Properties sold with the listing, reviews, and management relationship intact suffer less, which is part of what the turnkey premium is buying.
What causes a RevPAN estimate to be wrong?
| Error source | Direction | Typical magnitude |
|---|---|---|
| Mean instead of median on a right-skewed set | Overstates | 5% to 15% |
| Gross comp data against net expense stack | Overstates | 3% to 25% |
| Owner-blocked nights excluded from denominator | Overstates | 5% to 20% |
| Comp set matched on bedrooms but not amenities | Either | 10% to 30% |
| No first-year ownership discount applied | Overstates | 10% to 20% |
| Trailing period captured an unusual demand year | Either | 10% to 25% |
These stack. A model carrying three of them simultaneously can be 40 percent optimistic while every individual input looks defensible in isolation, which is why the errors are worth naming rather than hedging with a single blanket haircut.
Does a higher RevPAN always mean a better investment?
No, and treating it that way is the most common screening error after using ADR. RevPAN measures revenue productivity, not returns. The markets with the highest RevPAN relative to price are frequently the markets with the most active regulatory disputes, the thinnest insurance availability, or the weakest long-term rental floor underneath, and each of those shows up at exit rather than during the hold. RevPAN earns a property the right to proceed to the expense stack. It does not earn it a purchase.