Ask most restaurant owners what DoorDash or Uber Eats takes, and they’ll tell you the commission rate on their contract. Somewhere around 15% to 30%, depending on the tier they signed up for. That number is real. It’s also nowhere near what actually leaves your business.
Here’s an actual month, pulled from one pizzeria’s July statement. Not an industry average, not an estimate — one real payout report, read line by line.
Gross orders that month: $19,089. Amount the restaurant actually received: $10,307. That’s 54%. The platform kept 46%.
The contract said 24.7%.
Where the other 21% went
The commission is only the first line on the statement. Underneath it:
- Commission: $4,721 (24.7% — the number everyone quotes)
- Advertising fees: $1,424 (7.5%)
- Restaurant-funded customer discounts: $2,650 (13.9%)
- Error charges and adjustments: $168 (0.9%)
That third line is the one that surprises people. Those are the promotions the restaurant is paying for — the “buy one get one,” the “$5 off your order” — where the discount comes out of the restaurant’s side, not the platform’s. On this account it wasn’t a one-time misfiring promo. It ran between 8% and 19.5% of subtotal on every single payout, averaging about 14%. That’s structural. It had been running that way for months.
What that does to a single order
The delivery ticket was actually bigger than the walk-in ticket — about $40 versus $30. That sounds great until you run contribution per order:
- In-house order: about $20.66
- Delivery order: about $10.46
A delivery order with a 33% larger ticket produced half the contribution of somebody walking in the door. And on this account, 100% of the advertising spend was pointed at the channel making half as much per order.
To be fair to the channel: it still cleared roughly $4,990 in contribution that month. It was profitable. It was just the thinnest dollar in the building, and it was the one getting all the marketing money.
The advertising attribution problem
This is the part worth checking carefully on your own account.
The platform’s sponsored listing dashboard reported 436 orders and $17,265 in sales attributed to ads over a seven-week window. That works out to about 58% of all delivery volume supposedly created by advertising.
Then compare actual gross sales per day, month over month: $585, then $617, then $616. A 5% lift.
The attribution was claiming credit for more than half the channel, while the actual month-over-month movement was noise. Two numbers in the ad dashboard explain why: a 6.2% click-through rate, against a normal range of 2% to 5%, and a 25% click-to-order conversion. Those aren’t the numbers of strangers discovering a restaurant. Those are the numbers of people who already typed the restaurant’s name into the search bar and would have ordered anyway.
Net effect on that account: about $934 a month in ad spend producing about $239 in contribution. Roughly negative $695 a month, or $8,300 a year, to advertise to customers who already knew the name.
How to check your own
You don’t need a consultant to run this. You need one statement and twenty minutes.
- Pull a full payout-level statement, not the summary dashboard. The summary hides the discount lines.
- Add every deduction, not just commission. Ads, funded discounts, error charges, adjustments.
- Divide what you actually received by gross orders. That percentage is your real take rate. Compare it to what you thought it was.
- Check whether your delivery menu is marked up. Industry practice is 20% to 25% to offset commission. A surprising number of restaurants never set this and are selling at dine-in prices against a 46% haircut.
- Test your ad attribution. Compare gross sales per day across months, not the platform’s attributed revenue. If the daily average barely moved, the ads are billing you for demand you already had.
- Find every promotion with no end date. Discounts set up during a slow stretch have a way of running for a year.
One caveat on the numbers above: that was a month-over-month comparison during summer, with school out, and the seasonality wasn’t adjusted for. It was enough to act on. It wasn’t a controlled study, and yours won’t be either.
The point
Third-party delivery isn’t automatically a bad channel. On this account it was genuinely profitable and worth keeping. But an owner who believes they’re paying 25% when they’re paying 46% is making pricing, marketing, and staffing decisions on a number that’s off by nearly half — and when a restaurant is already tight, that gap is often exactly the size of the problem.
Read the statement. The answer is in there.
If you want a second set of eyes on yours, book a time.