Hotel reporting software: which numbers actually matter

Occupancy and ADR mislead on their own.

Take a 40-room hotel across a 30-night month. That is 1,200 available room-nights, and it is the denominator behind almost everything worth knowing.

In March the hotel sells 960 of them at an average of £90. That is 80% occupancy, £86,400 of room revenue and a RevPAR of £72. In April it sells 720 at an average of £130. That is 60% occupancy—a figure that will worry an owner who asks only about occupancy—£93,600 of room revenue and a RevPAR of £78. April was quieter, looked worse on the number most people ask for first, and earned £7,200 more.

RevPAR, revenue per available room, is occupancy multiplied by ADR. That is why it is the first figure worth having: it cannot be flattered by discounting into a full house, and it cannot be flattered by holding rate into an empty one. Either lever on its own moves one of its two components and gives the other away.

ADR carries its own trap. It is an average of what was sold, never of what was offered. A month with thirty rooms sold at £200 and 690 sold at £127 produces almost the same ADR as a flat month at £130, and the two months need completely different responses. An average that never moves is not evidence of stability; it is evidence that nobody has looked underneath it.

RevPAR does not know what the room cost to sell.

Two hotels finish April with an identical RevPAR of £78 and identical room revenue of £93,600. The first sells 70% of that through commissioned channels; the second sells 30%. At 15% commission, the first pays £9,828 and the second £4,212. The gap is £5,616 in a single month—about 6% of room revenue—and it appears nowhere on a report that stops at RevPAR.

This is not an argument for switching distribution off. Reach costs money and frequently earns it, and a hotel that fills a soft February through a channel has bought something real. It is an argument for a second column. Once the cost of the sale is visible, a channel decision becomes a comparison rather than an argument about principle.

Payment fees, booking-engine charges and the cost of a cancelled-and-rebooked stay belong in the same column. None of them is large on its own; together they are the difference between a month that looked fine and a month that was.

The industry already has a vocabulary for this. USALI, the lodging industry’s standard for financial and operating reporting, has a mandatory adoption date of 1 January 2026 for its 12th revised edition. A 40-room independent does not need to adopt a full accounting standard. It does need its definitions to stay still from one month to the next, and borrowing them from a published standard costs less than inventing them.

Last month is history. On the books is the decision.

Almost every hotel report describes a period that has already finished. Nothing on it can be changed. The one figure an operator can still act on is what is already sold for nights that have not happened yet.

On 1 August a hotel holds 340 room-nights for September. On 1 August last year it held 400 for the September that followed. The useful number is not last September’s final occupancy. It is the 60 room-nights it is behind at the same lead time, and it is available now rather than in October. That figure is what justifies opening a rate, releasing an allocation or spending on a campaign—while there is still a September to sell.

Two words describe this, and both are worth using precisely. Pickup is what was added over a period for a future period. Pace is where the hotel stands compared with the same point last year. Pickup tells you the direction; pace tells you whether the direction is good enough.

The catch is that pace only works if somebody recorded the position at the same point last year. Most hotels discover in their first year of trying that nobody did. Some systems can reconstruct it from booking dates; many cannot, and cancellations usually make the reconstruction unreliable. A single line written down each week costs nothing and settles the question permanently.

Most reporting arguments are definition arguments.

The fastest way to make a hotel distrust its own figures is to let two systems answer the same question differently, both of them correctly.

Take occupancy in that 40-room hotel again. Four rooms are out of order for the whole month, so 120 of the 1,200 room-nights cannot be sold. The hotel sells 885. Against 1,200 available room-nights that is 73.8%. Against the 1,080 it could actually sell, it is 81.9%. Eight points of difference, one month, no error anywhere—and two people who will each defend their number in a meeting.

The same applies to VAT. An ADR quoted gross is not comparable with an ADR quoted net, and a hotel that mixes the two across systems will spend a year arguing about rate positioning it never actually lost. HMRC’s VAT Notice 709/3 also covers deposits and cancellations: a deposit taken and retained when a guest cancels is not automatically outside the scope of VAT, and where that money lands in your reporting is a decision somebody has to make once, in writing.

The CMA’s price transparency guidance pushes on the same seam from the guest’s side, covering mandatory fees, taxes and charges within the price shown. If a mandatory charge appears in the total the guest sees, decide once whether it is room revenue or other income. Otherwise two reports will disagree with each other and with the guest’s receipt.

None of the questions below has a right answer. They have a hotel’s answer, and the value comes from writing it beside the number rather than from choosing well.