Which Channel Carries the Margin After the Fee Migration

When the fee model changes, the ranking of channels changes with it. The arithmetic is one subtraction, done per channel.

Last updated: November 1, 2026

The migration didn't change the prices on your listing. It changed what arrives after the platform takes its cut, and it changed that unevenly. Which channel now carries the best margin isn't a fact you can look up, and it isn't the same answer for two hosts on the same street. It's a subtraction you run once per channel, on your own prices.

Key Takeaways

• One identity, four runs. Net equals listed price minus the platform fee. The fee differs per channel. The subtraction doesn't.
• The base matters more than the label. A fee applied to a total that includes cleaning isn't the same fee as one applied to the nightly amount, even when the two look similar on a settings page.
• A ranking is an output, not an input. Anyone who tells you which channel wins without seeing your prices is describing their own portfolio.
• Compare like with like. Same unit, same dates, same currency, same cancellation terms. A net from one channel against a gross from another tells you nothing.
• Re-read the numbers when a fee notice lands. A comparison built on last season's model is a guess wearing a spreadsheet.

What actually changed in the fee model

Two things moved at once, and hosts tend to notice only the first. The fee changed which side of the booking it's charged to. It also changed what it's charged on.

The first change is visible on a guest's screen, because the total a guest sees is part of how the listing competes. The second is visible only on your own statement, because it decides how much of the listed price survives to the end. A host who tracks only the guest-facing total will believe nothing happened to their revenue per night.

The useful way to hold both at once is to stop thinking about price and start thinking about two numbers per booking: what the guest agreed to pay, and what reaches you. Before the migration those two moved together predictably. They don't any more, because each channel now takes its cut at a different point in the chain.

What didn't change is just as important. Your costs didn't move. The cleaning bill is the same, the linen is the same, the night you can't sell twice is the same. That's why a fee change is never a pricing problem on its own. It's a margin problem, and margin is a subtraction.

Net is one subtraction

Write it once and it holds for every channel you sell on.

Net equals listed price minus the platform fee.

That's the whole formula. Everything else is a decision about which price and which fee go into it.

The listed price is the one you set for that night, on that rate plan, for that channel. Not the average across the month, and not the price on your best-performing weekend. Pick a real date, one you'd actually sell, and use the price that applies to it.

The platform fee is the amount the channel takes for that booking, as the channel itself states it, on the base it states. This is where hosts go wrong, and they go wrong in the same way every time: they use a number from memory, or a number from a forum, or a number from the fee model that applied before the migration. The only fee worth putting in the subtraction is the one on your own statement for a booking like the one you're comparing.

Then run it four times. Airbnb, Booking.com, Agoda, Trip.com. Same unit, same dates, same currency. Four subtractions, one per channel, and four net figures that can finally be compared.

There's a version of this that's easier to maintain, and it's worth the setup. A rate plan holds the nightly price for a set of dates, so the listed price in the formula is already written down per plan rather than remembered. localsbnb.com sets nightly prices per rate plan, and shows each channel's own rate and status beside the night it applies to. Across the four connected platforms, that puts the first term of the subtraction in the same view as the channel it belongs to.

What the subtraction can't tell you is whether a booking will happen. It's an arithmetic identity, not a forecast. Run it on a date you'd sell and you learn what that night is worth per channel. Run it on a date nobody wants and you learn the same thing about a night you won't sell. Both are useful. Neither is a projection.

Why the ranking moves

A ranking moves because the same change lands on different bases. That's the whole mechanism, and it's worth sitting with, because it explains why two hosts can run the same subtraction and get different orders.

Start with the base. Where a fee applies to a total that includes the cleaning charge, the cleaning charge is part of what the fee is taken from. Where a fee applies to the nightly amount alone, it isn't. A host with a high cleaning charge and short stays feels the difference more than a host with a low charge and long stays, on the same channel, with the same listed price.

Then the side. A fee charged to the guest changes what the listing looks like in a search result. A fee charged to the host changes what you keep. Both can end up costing the same amount, and they don't behave the same way, because one competes on the guest's screen and the other simply leaves the building.

Then the shape of your calendar. A channel that brings you long stays in the low season is exposed to a fee on the nightly amount. A channel that brings you short weekend bookings is exposed to a fee on a total that carries a fixed charge per stay. Same host, same unit, two different exposures.

Put those three together and the order can change without anyone's price changing. That's why this article stops at the formula. Telling you which channel comes out ahead would mean publishing an order built on one portfolio's cleaning charge, one portfolio's season mix and one portfolio's dates. Yours won't match, and the number you'd act on would be somebody else's.

The honest version is duller and more useful. Run the subtraction, write the four figures down, and let the order fall where it falls this quarter. Then ask the question that actually changes anything: if a channel has slipped, is that a price problem, a minimum-stay problem, or a channel you should use less.

Re-reading your own numbers

Do this once a quarter, and again within a week of any fee notice. It takes an evening and it replaces a year of guessing.

Pick one representative week, not your best one. A week in the middle of a normal month, on a unit you know well. Write the dates down, because next quarter you'll want the same week and the calendar will have moved.

Then, for each channel, fill four lines: the listed price for that date, the fee the channel states for a booking of that length, the net figure, and the date you checked it. Four channels, sixteen lines. If a line has a question mark instead of a number, that's the finding for this quarter, and it's worth more than the other fifteen.

Then decide, and write the decision down. A price change, a minimum-stay change, or no change at all with a date to look again. The third option is a real one. A channel that slipped in the order but still pays for the night doesn't need a reaction, it needs a note.

One habit makes the next quarter cheaper. Keep the fee notices, and copy the numbers on them into the same file as the table. When the model moves again, the before and the after sit in one place.

FAQ

Is the channel with the lowest fee always the one with the highest net?

No. The fee is one term of the subtraction and the listed price is the other, and a channel where you set a higher price can end up ahead of one where the fee looks smaller. The order only means something when both figures come from the same dates, the same unit and the same currency.

Should I raise my listed price after the migration?

That depends on what the subtraction says for your unit, and it isn't the only lever. A longer minimum stay reduces how often a fixed per-stay charge applies. Fewer channels on the tightest unit reduces exposure. Change one thing at a time, on one unit, so you can tell which change did what.

How often should I re-run this?

Quarterly, plus once within a week of any fee notice. A comparison built on a model that has since changed isn't wrong in a way you'll notice. It's quietly off, and it stays off until you rebuild it from statements rather than memory.

One identity, four runs, one evening a quarter. localsbnb.com keeps nightly prices per rate plan and each channel's rate beside the night it applies to, which is where the first term of the subtraction already lives.

---

This article is general guidance for hosts and isn't legal, tax or insurance advice. Platform fees, fee bases and payout terms differ by platform, by market and by account, and they change; confirm the current terms of each channel you sell on before you price against them.