Which booking channel actually makes you the most money?

The channel with the highest rate is rarely the one that leaves the most behind. A room sold at €181 through an agent taking 15% leaves €153.85 in the hotel. The same room sold at €172 on your own website leaves €170.84. Your rate report ranks the first night as the better one, and your bank account disagrees.

The figure that settles it is net ADR by channel: the rate a booking achieved, less the cost that booking caused, per roomnight. It is a simple calculation and almost no independent hotel can produce it, because it needs the property management system (PMS) and the ledger to agree on what a channel is.

This is Part 4 of USALI for independent hotels, from the general ledger up. Part 1 made the case that a standard accounting package produces an accurate P&L but not a departmental one. Part 2 described the shape the standard gives a hotel’s accounts. Part 3 covered the chart of accounts, and ended with a ledger that can finally say what each channel costs. This chapter is about what to do with that, and it is the first of the chapters that take the departments one at a time.

What does a booking actually leave behind?

Take an illustrative 88-room hotel over twelve months. It sold 24,000 roomnights against 32,120 available, so occupancy ran at 75%. Blended gross ADR was €165.

It sells through six channels. By volume: Agent A 7,400 roomnights, its own website 6,200, groups and meeting planners 3,100, corporate and GDS 2,900, Agent B 2,600 and wholesale 1,800. Agent A invoices the hotel for commission. Agent B collects from the guest and passes on a net rate.

Here is what each one leaves behind, with cost as the Rooms department carries it:

Read the two rate columns side by side and the order changes completely. On gross ADR, Agent A is the hotel’s best channel and its own website is second. On net ADR, Agent A falls from first to fourth, and the website goes top by nearly €17 a night.

Three things follow from that table, and each of them is worth more than the ranking itself.

• Total distribution cost, €261,878. That is the difference between €3,959,200 of gross rooms revenue and €3,697,322 net of channel cost. That is 6.6% of rooms revenue, sitting inside the Rooms department.
• Net RevPAR €115.11, against a gross RevPAR of €123.26. Both are true. Only one of them is money the hotel keeps, and only the second one appears in most monthly reporting.
• Average cost of a sold night, €10.91. That is the gap between the two rate columns, and it is what the hotel pays, on average, to have someone else sell the room.

None of this requires a new system. It requires the rooms sold by channel that the PMS already holds, and the commission and fee lines that Part 3 put into the ledger.

Why do two agents charging the same produce different accounts?

Because the money arrives by two different routes, and only one of them creates an expense.

• The agent model. The guest pays the hotel. The agent invoices commission afterwards. Rooms revenue is the full €181, the commission is a real cost in the Rooms department, and net ADR is a subtraction you perform.
• The merchant or net rate model. The agent takes the guest’s money and passes on an agreed net rate. Nothing is invoiced. Revenue is recorded at the €157 received, there is no commission line anywhere, and net ADR is simply the rate.

Part 3 set the accounting rule: you never gross up a net rate to make the two look alike, because inventing a commission that was never paid overstates revenue, ADR, RevPAR and expenses all at once. The measurement rule is the mirror of it. You cannot compare a gross rate with a net rate. Put both on a net basis and the comparison is honest in either direction.

That matters commercially, because the two models make the same arrangement look completely different in a rate report. Agent B appears to be a weak channel at €157 against Agent A’s €181. After cost, Agent B is the better of the two. A hotel that judges its agents on rate alone will negotiate hard with the wrong one.

Two cautions from the same ground:

• One agent can sell both ways, sometimes on the same property in the same month. Record them as separate arrangements. A euro of revenue means something different in each, and averaging them produces a rate that describes neither.
• A net rate is not a discount you chose. It is a margin taken inside a rate you agreed. The number to negotiate is the net rate itself, because that is the only figure the hotel ever sees.

Is a direct booking really free?

No, and the way the standard places costs can make it look that way.

USALI puts a cost where the thing you bought belongs, not where the booking came from. A per-booking fee charged by the system that carried the reservation sits in the Rooms department. Paid search and metasearch campaigns are exposure rather than bookings, so they belong in Sales and Marketing. The cost of taking a card sits in Administrative and General, on every channel where the hotel takes the payment rather than the agent. All three are correct, all three follow from Part 3’s test of what you paid for rather than who you paid, and the effect is that the Rooms department sees almost none of what a direct booking costs.

That is not an argument for moving the costs. It is an argument for two views of the same channel:

Direct still wins, and by a little less than the rate comparison suggests: its lead over the commission agent narrows from €16.99 a night to €14.11, and over the next best channel from €9.24 to €6.18. What changes is how much of the argument for driving direct business survives contact with its own costs, and on these figures most of it does.

Say which basis you are using, every time. The departmental figure is the one that reconciles to the accounts and the one another hotel could ever be compared against. The acquisition figure is the one that answers a commercial question. Quoting one and calling it the other is how a channel review ends in an argument nobody can settle.

What is a change in channel mix actually worth?

This is the calculation that turns the table into a decision.

Moving 500 roomnights from Agent A to the hotel’s own website is worth €16.99 a night on the departmental basis, or €8,495 a year. On the acquisition basis, after the paid search and card cost that come with winning that business directly, it is €14.11 a night, or €7,055.

Both figures are real and neither is a forecast. They describe what would have happened to last year’s trading if the mix had been different, which is a far safer claim than a projection.

Three qualifications, because a mix shift is not free money:

• The nights have to exist. If Agent A’s guest would not have found the hotel otherwise, moving the booking does not move the revenue; it removes it. The calculation values a substitution, not a new customer.
• Displacement runs the other way too. A high-cost channel that fills a Tuesday in November is worth more than an empty room, whatever its net ADR. Channel decisions belong to periods, not to the year.
• Mix is worth measuring by nights and by net revenue. Agent A is 31% of roomnights and 31% of net rooms revenue on these figures. When those two percentages diverge, the difference is the story.

Channel mix is a profit decision before it is a distribution one. Made on gross rates alone, it is a decision made on the wrong number.

Why does your ADR not match the PMS?

Ask for the ADR and you will often get two answers, from two systems, both correct. The differences are almost always definitional, and they recur in the same places when we reconcile a hotel’s PMS to its ledger:

• What is inside the rate. A package that includes dinner is one payment and two departments. If the PMS reports the whole amount as room revenue and the ledger splits it, ADR will disagree by the food element on every package night.
• What counts as a room sold. House use, complimentary rooms and a room held for a long-stay contractor are treated differently by different systems, and the denominator moves even when nothing was sold.
• Timing. Deposits taken in one period, no-shows charged after departure and city ledger balances settled in arrears all land in different months in the two systems.
• Tax. One figure is usually net of VAT and the other is not, which is a 9% or 13.5% difference in Ireland depending on the supply, and enough on its own to make two correct systems disagree.

A VP of Accounting at an international group told a HOSPA webinar in February 2025 that aligning the PMS with the accounts on segmentation and channel is one of the more delicate parts of a USALI migration. That matches what we find. The fix is not a reconciliation each month; it is one agreed definition per channel, written down, used by both systems.

And the reason to bother is that the derived figures inherit every one of these differences. RevPAR is occupancy multiplied by ADR, so a rooms-sold definition that drifts moves RevPAR without anything happening in the hotel. The hotel valuation firm HVS puts the correlation between GOPPAR and a hotel’s value at between 85% and 90%, against roughly 70% to 75% for RevPAR. Rate and occupancy describe the top of the Rooms schedule. What the owner is paid on is what survives to the bottom of it.

What can you measure from month one?

More than most people expect, and less than everything.

• Available from the start: roomnights and gross rate by channel, for as far back as the PMS holds them. Channel mix is a historic series on day one.
• Available from the month the ledger carried the detail: the net cost of each channel, and therefore net ADR. A new account’s history starts when it is created, which is the one real penalty in the whole exercise.
• Never available: a net ADR history reconstructed by spreading last year’s single commission figure across last year’s channels. It would look like a measurement and it would be an allocation. An estimate that looks like a measurement is more dangerous than a gap.

The 12th edition of USALI, which hotels have been adopting since January 2026, adds an optional schedule for rooms revenue by booking channel. It is optional, and it covers revenue rather than cost, so it does not produce net ADR on its own. It does mean the question this chapter asks is now one the standard expects a hotel to answer.

The SSOT Platform is built around that division. The FC uploads the GL transactions and the monthly statistics, rooms sold and the rest, and the management accounts come back in the USALI layout with every line drillable to the transactions behind it. Net ADR by channel is produced going forward, from the month the ledger carries the detail. Historic gross ADR by channel comes from the PMS. The historic net split is not offered, because the data does not support it.

A number you can act on

Gross ADR tells you what you charged. Net ADR tells you what the hotel kept, and the two can rank your channels in a different order, as they do for the hotel in the table above.

Three questions to take back to your own ledger:

• If the owner asked which channel earned the most per room last year, would you answer with a rate or with a rate net of what it cost?
• Do your rate report and your accounts use the same definition of a room sold, and can anyone in the hotel name the difference?
• When channel mix moves next quarter, will you be able to say what it was worth?

USALI (the Uniform System of Accounts for the Lodging Industry) is published by HFTP. SSOT Analytics is independent of HFTP. This series explains the concepts in our own words; for the authoritative text, buy the 12th edition at usali.hftp.org, or in the UK and Ireland via hospa.org. There is no substitute for reading the book.

If the first question has an uncomfortable answer, that is the place to start. Talk to us about making your hotel’s reporting benchmark-ready, or follow the series and do it yourself.

Colin Donovan has spent 30 years in hotel finance, as financial controller, general manager and managing director. He is the founder of SSOT Analytics, which builds USALI-aligned financial reporting for independent hotels.