What the Platform Takes
The pitch from the delivery platforms is not complicated. You open a new revenue channel. Orders reach people who would never pass your door. The kitchen earns something on the nights the room is quiet.
The catch is in the commission.
On a typical order through one of the major UK delivery apps, the kitchen sends somewhere between a quarter and a third of the order value back to the platform before the food is out the door. That number comes off the top. Before ingredients. Before the portion of wages the order absorbs. Before the rent it earns its share of. The platform is first in the queue, every order, every night.
The maths that does not add up
A kitchen running on a net food margin in the single digits has no room to send a third of revenue somewhere else. Most cannot make it work at standard menu prices. So the industry found a workaround: price the delivery menu higher than the dine-in list. The same dish, twenty or thirty percent dearer on the app.
There is a logic to it. The customer ordering delivery is not sitting at the table, so perhaps they expect to pay for the convenience. The problem is that the restaurant absorbs that framing. Guests who dine in and order delivery learn quickly that the prices differ, and the message that lands is not "delivery is priced for convenience." It is "this kitchen does not have consistent prices."
Some venues price it cleanly and communicate the difference. Most manage an awkward gap. None of them found a version that does not involve a compromise somewhere.
The platform always gets paid
The thing about delivery commission worth sitting with is this: the platform gets paid regardless of whether the kitchen does.
A slow Thursday with twelve covers still sends its commission. A wet February with half the usual trade still cuts the platform in on every order that came through. The platform does not share in the risk of a bad month. It takes its percentage from revenue you generated and earns its cut before you reach the margin that determines whether the night was worth running.
Operators who lean into delivery as their main growth channel often find the order count is real and the profitability is not. The kitchen works harder, the courier takes another order, and the margin does not move.
The fixed cost dressed as a percentage
What delivery commission actually is, once you strip the growth framing away, is a fixed overhead attached to a revenue line. Not fixed in pounds, but fixed in the sense that it does not go away when trading is hard. It scales with revenue, not with profit.
Most hospitality fixed costs are at least predictable. Rent is the same every month. Energy is the same per unit. Business rates were calculated once and landed as a bill. Commission is different: it grows with turnover and shrinks with nothing. A good month and a bad month both pay the platform first.
A different kind of deal
Halfseat does not take a percentage of the food. The food at a Halfseat booking is already at half price, because it is on a seat that was going to earn nothing that night regardless. What Halfseat takes is a small booking fee from the diner, and three pounds of that goes directly to the venue. Every drink, from the first glass to the last round at the bar, stays at full price. The margin on drinks is the margin that keeps a hospitality business running, and it stays intact.
The delivery model extracts a percentage from revenue you already had, on nights you were already trading. Halfseat recovers something from a seat that had none. Those are different trades.
For a kitchen already managing twenty percent VAT on every plate, a rising wage floor, business rates and food inflation that has not fully settled, the question of what percentage leaves the building through an intermediary is not abstract. It is a real number landing every month, on a margin that has nowhere to absorb it.
The platform always gets paid. The question every operator running a delivery partnership is sitting with, somewhere around month four, is whether the kitchen does too.