Two identical houses on the same street. In one season the first sells 190 nights, the second 160. The first takes in €600 more. At the end of the year, the second has nearly €3,260 more in its pocket.
It isn't an accounting trick. It is the simplest and least watched fact in short-term rentals: you pay for the arrival, not the night. Every check-in brings a cleaning, a change of linen, an hour of messages. The first house had 95 two-night arrivals, the second 40 four-night ones. The first house has the prettier calendar. The second has the far prettier bank balance.
A Fast Fleet That Didn't Know Where It Was
In 1707 four ships of the British fleet ran onto the rocks of the Isles of Scilly, and more than a thousand sailors died. The captains knew exactly how fast they were going and in which direction. What they didn't know was where they were: back then, nobody could work out longitude at sea. John Harrison, a carpenter turned clockmaker, solved it with a clock that kept the time of the home port on board. The experts distrusted him for years: the instrument worked, but it was too simple to look serious.
In short-term rentals, occupancy is the ship's speed: it tells you how much you're moving, not where you're heading. The chronometer is a different number, how much you keep for every night you sell, compared with your own previous year and with the market around you. The technical name is financial benchmarking. In practice it means you stop staring at the calendar and start reading the chronometer.
Why Almost Nobody Calculates It
A full calendar can be seen, photographed, talked about. At industry events that's how people introduce themselves: "we're always full". Nobody adds "and I have no idea what I keep", because it doesn't land quite the same over a drink.
Net per night, on the other hand, appears nowhere: not in the portal dashboard, not in the bank statement, not in the property management software, unless you build it yourself. And building it is boring. So people look at the number they can see and, the moment a gap appears in the calendar, they drop the price.
The reflex is understandable: an empty night feels like a clean loss. But a night sold badly can cost more than an empty one. You pay for the cleaning, the commission, the electricity, you wear out a little piece of the house, and you take in barely more than you spent. In the house that fills up at any cost, 95 arrivals in one season mean 95 cleanings, 95 linen changes, 95 times someone asks how to switch on the water heater.
The Three Numbers on the Chronometer
Three are enough to know where you are. Only three, but they have to be read together.
Net per night sold. Take your revenue, subtract portal commissions, payment costs, cleaning and variable expenses, and divide by the nights you sold. It's the number that tells you whether you're really growing or just working harder.
The cost of opening the door. This is the heart of the matter: cleaning, linen, welcome kit and time spent with the guest are paid per arrival; energy and wear on the house per night. Keeping them apart changes everything, because it explains why a house with longer stays earns more even while selling fewer nights. And it gives you your floor price: below that figure, the empty night wins.
The weight of each channel. On Booking.com the base commission usually sits around 15%, depending on country and contract; add payments and visibility programmes and, for many hosts, the real cost reaches 18–23%. For a guest who would never have found you on their own, that's a fair price: the portal has an audience none of us can reach alone. The point is knowing how much each channel weighs on the total. If 90% of your bookings come through a single door, every rule change hits you in full. Measuring it helps you decide where to invest, not to pick a fight with anyone.
The two houses, line by line
Two identical houses, a 200-night season, the same costs. Example figures, not market averages.
House A: full at any cost
- 190 nights sold (95% occupancy) at €100 a night, in 95 two-night stays
- Gross revenue: €19,000
- Portals, commission and payments (18%): −€3,420
- Cleaning and linen (€60 × 95 arrivals): −€5,700
- Energy, utilities, maintenance reserve (€15 × 190 nights): −€2,850
- What you keep: €7,030, or €37 per night sold
House B: fewer arrivals, longer stays
- 160 nights sold (80% occupancy) at €115 a night, in 40 four-night stays
- Gross revenue: €18,400
- Portals, commission and payments (18%): −€3,312
- Cleaning and linen (€60 × 40 arrivals): −€2,400
- Energy, utilities, maintenance reserve (€15 × 160 nights): −€2,400
- What you keep: €10,288, or €64 per night sold
House B takes in €600 less and keeps €3,258 more. Run the numbers with your own figures: the proportion usually holds.
Why RevPAR rewards the wrong house (for the professionals)
RevPAR (revenue per available night) is the industry's favourite number, and rightly so: it combines rate and occupancy. But it looks at gross. In our example: House A €95, House B €92. A wins.
Hotels correct it with net RevPAR, which takes out distribution costs: A €77.90, B €75.44. A still wins.
In short-term rentals that isn't enough, because the heaviest cost isn't the commission: it's the arrival. A hotel room is turned around in twenty minutes with a trolley; an apartment needs a team, a car and rented linen. Take out the per-arrival and per-night costs as well and you get net per available night: A €35.15, B €51.44. Now B wins, and by a long way.
If the report you send the owner stops at occupancy and RevPAR, it is rewarding the wrong house.
2026 Makes All This Less Optional
Operating costs aren't coming down, more and more cities are regulating short-term rentals with a firmer hand and, since 20 May 2026, a European regulation requires platforms to share short-term rental data with the authorities: nights sold, guests, registration numbers. In a more transparent market, whoever doesn't know their own numbers will be the only one who doesn't.
Numbers Only Work If Someone Understands Them
An exact number delivered badly is still just a number. The owner who receives, at the end of the month, a sheet full of acronyms and not a single word has data in hand, not answers. And people who don't understand, sooner or later, stop trusting.
The monthly report is where numbers meet people. It isn't there to prove how hard you worked, but to tell what happened to the house: why the average rate went up, why an empty week was a choice and not a hole, what the boiler cost and what it saved. The same three lines every month: net per night, cost per arrival, weight of the channels.
Then there's the uncomfortable side, the one that matters most: saying when a month went badly. Hiding a dip for a quarter buys peace and quiet; explaining it straight away, with the numbers and a plan, buys trust. Peace and quiet ends with the accountant's first phone call. Trust holds a relationship together for years.
What to Do Tomorrow
Take the season you've just closed and redo the two-house sum with your own numbers: revenue, minus portals, minus cleaning (arrivals × the cost of one cleaning), minus variable expenses, divided by nights sold. Then do it again for the year before. An afternoon and your list of bookings are all it takes.
If net per night is going up, good: you know what's working. If it's going down while the calendar was filling up, you've found where the money goes. The first levers to try cost little: a floor price you never sell below, a minimum stay of a few nights in the periods when demand is there anyway, a small perk for guests who stay longer.
And a question for anyone managing other people's homes: do you know what one arrival costs you, separate from the night? If the answer is no, it's the first number to look for on Monday morning.
Harrison didn't win because he was the fastest, but because he had a fixed point. First the chronometer, then the sails.
If you'd like to run the numbers with us, get in touch: the first conversation is free and comes with no strings attached.
