How hotels decide tonight’s rate, and three of their tools you can use for free

What happened

Nothing new, which is the point. Hotels have a discipline called revenue management, and the large ones run software for it. It came over from the airlines in the 1980s, and there is a deep academic literature behind it, including at Cornell's hotel school.

It turns on one question: is this the right price today, or would waiting get more? A room that does not sell tonight is gone — you cannot hold it over. That makes the timing of the decision the decision. One unit or five hundred, the problem has the same shape.

What it means for you

Three of the tools cost nothing and need no software.

1. Look at RevPAR, not your nightly rate

Hotels do not judge themselves on average daily rate alone. They use RevPAR = average rate × occupancy. It catches the case where you raised the rate, lost the bookings, and came out behind.

Three months of the same listing:

	Average rate	Nights sold	RevPAR (per 30 days)	Revenue
September	₩100,000	25	₩83,000	₩2.50m
October	₩150,000	15	₩75,000	₩2.25m
November	₩120,000	21	₩84,000	₩2.52m

October has the highest rate and the lowest revenue. If you only track your nightly rate, October is the month you remember as a good one.

One trap: RevPAR is revenue, not profit. More nights sold means more cleaning, laundry, consumables and messages. To see what you actually keep, run the number on payout minus what cleaning really costs you. A cleaning fee collected from the guest covers a cost; it is not margin.

2. Pickup — count when the nights filled, not just whether they did

Hotels record how much is on the books at 30, 14 and 7 days out. The point is not the final number but the speed of filling. The final number arrives when the month is over; the speed arrives while you can still do something.

One line a week next to your calendar is enough:

Nov 24 → December: 8 of 31 nights booked (26%)
Dec 01 → December: 13 of 31 (42%)
Dec 08 → December: 19 of 31 (61%)

What if you have no last year to compare against? First-year hosts get stuck here. Three substitutes:

• Compare weeks within the same month. If the second week of December is 61% and the third is 20%, that week is priced wrong or blocked by a condition
• Change the price once and watch the speed. Drop 10% for two weeks and see whether pickup moves. If you have no baseline, make one
• Count the days left. Seven days out and half empty is already the answer

3. Sell the same room at different prices — by attaching conditions

Hotels charge several prices for one room because each price carries a condition. Non-refundable is cheaper, three-night minimum is cheaper, booked two months ahead is cheaper. The guest chooses, so nobody calls it unfair.

The hotel toolkit does not transfer to Airbnb one to one. What a host can actually reach is narrower:

The hotel move	What you have on Airbnb	What it costs you
Non-refundable rate	Nothing equivalent — you pick from set cancellation policies	—
Minimum stay	Per-date minimum nights	Filters out short bookings entirely
Advance-purchase rate	Early bird discount	Also discounts nights that would have filled anyway
Long-stay rate	Weekly and monthly discounts	Cuts a full week in one step

How to choose: lowering your base rate applies to every booking. Attaching a condition applies only to guests who accept the condition. If you are worried about a slow season, try the conditional levers before you move the base rate.

What is different when you have one unit

Hotel revenue management assumes inventory — many rooms. A 200-room hotel can sell some cheap and some dear on the same night. You have one.

• With nothing to spread across, it is all or nothing. If tonight does not sell at this price, tonight earns zero. That is why pickup has to be read earlier
• Your sample is small, so one month proves little. Three nights' difference across thirty can be noise. Stack two or three months before concluding anything

Three things to do this week

1. Work out RevPAR for the last three months. Payout for the month ÷ nights in the month; subtract real cleaning cost for a truer number. At least one month will read differently than it did on rate alone.
2. Start logging pickup. One line, same day each week. The three substitutes above work even with no history.
3. Add exactly one conditional discount. A minimum stay or an early bird. Attaching a condition protects your average rate better than cutting the base.

---

The three-month table is an illustration of the arithmetic, not a real listing's results. RevPAR, pickup and rate fencing are standard revenue-management concepts, not any vendor's method. Checked September 23, 2026.