A plain-English breakdown of what marketplace commission really costs once every fee stacks, plus the napkin arithmetic to work out your own number instead of trusting a rate card.
Short answer: a 30% marketplace commission is not 30% of your profit. It is roughly 30% of the menu price, taken before you have paid for a single ingredient. On a $32 delivery order at a 28% blended rate, $8.96 leaves the top. You receive $23.04, you still owe roughly $9.60 in food cost and $10.24 in labor, and you keep about $3.20. That is the real shape of the number. Below is the arithmetic, step by step, so you can work out your own figure on a napkin in fifteen minutes using your payout statements instead of anyone's headline percentage. Every example here is labelled as an example.
A marketplace is an app where the diner finds you inside somebody else's storefront. DoorDash, Uber Eats and Grubhub are marketplaces. They bring you the customer, and they keep the customer.
When someone says "we are only paying 15%", they are quoting the base. Almost nobody's blended number is 15%.
Three inputs. That is the whole thing.
Monthly commission = orders per month x average ticket x rate
Multiply by twelve for the annual figure. Most owners have never done this multiplication, which is why the number stays fuzzy.
This is an example, not a customer. A single-location independent doing 700 marketplace orders a month at a $32 average ticket, so $22,400 a month in marketplace food sales. It sits on a 25% base tier and thinks of itself as a 25% shop.
Annualized, this example restaurant pays $75,264 to $88,704 a year to be found inside somebody else's app. The owner would have told you they pay 25%.
Again, an example. One $32 delivery order at a 28% blended rate, walked all the way down. Food cost set at 30% of menu price and labor at 32%, the middle of the range most independents run. Use your own percentages.
The commission on that order, $8.96, is 2.8 times what the order leaves behind. You bought the food, paid the labor and carried the risk, and the app took almost three times what you kept.
Run the same $32 order on your own site with no commission and you keep $32.00 instead of $23.04. Same food, same labor. Left over: $12.16 instead of $3.20. Nothing about the cooking changed. Only the toll changed.
Rate cards describe a plan. Payout statements describe reality. Pull the last three full months from each app and find two figures: gross food sales (what diners paid for food, before any deduction) and net deposits (what landed in your bank). Then:
Your real blended rate = 1 - (net deposits / gross food sales)
Your real commission rate is one minus your net deposits divided by your gross food sales, and it is almost never the number on the rate card. Average three months, because a month with a promotion running skews high and a quiet month skews low. Exclude diner-paid delivery fees and tips from both figures or the ratio comes out wrong. Food compared to food.
Example: $22,400 gross food sales, $16,128 net deposits. 16,128 divided by 22,400 is 0.72. One minus 0.72 is 0.28, a 28% blended rate. Now you have your number, not a brochure's.
Commission scales with your success. A flat monthly fee does not. That is the whole argument for direct ordering, and it is worth one more calculation.
Zayos (also written Zay-OS) is commission-free direct ordering. Diners order on your own branded site, orders land on the same kitchen tablet you already run for the delivery apps, and pricing is $399, $499 or $599 per location per month, no setup fee, month to month, no contract. You keep 100% of food revenue and 100% of tips. The checkout fee is paid by the diner, never by you: $0.99 pickup, $0.99 delivery, $0 dine-in, 10% on catering.
So: how many orders do I move off the apps before the subscription pays for itself?
Break-even orders = monthly fee / (average ticket x blended rate)
Same example numbers: $499 divided by ($32 x 0.28) is about 56 orders a month. Under two a day. Every order past that is margin coming back to you.
Two honest caveats. Card processing exists on every channel, including your own front counter, and the processor charges it, not us. And direct delivery dispatches through a fleet at a flat per-delivery fee instead of a percentage. Both are worked through in the direct-ordering payback post.
So, plainly:
Past that we can offer a model, not a result. Run the formulas above at scale and recoverable commission lands between $48,000 and $300,000 per location per year, at 3,000 to 12,000 orders a month. That is arithmetic, labelled as arithmetic. Check it against your own statements.
Keep the apps for discovery. They are still the biggest new-customer channel in food. Just stop paying that fee on the regular who already knows your name, your hours and their usual order. The levers that move those regulars are in the commission-reduction playbook, with the Florida version in the commission-free ordering guide.
Real as a base rate on the top delivery tiers, and blended cost usually lands between 25% and 35% once sponsored listings and absorbed refunds pile on. Check yours with the payout-statement formula above. Fifteen minutes.
No, and you probably should not. Treat marketplace commission as an advertising cost for finding brand-new customers, because that is what it is. The money is in moving repeat customers to direct, where the toll is zero.
$399, $499 or $599 per location per month, no setup fee, month to month. You keep 100% of food revenue and 100% of tips. The diner pays $0.99 on pickup, $0.99 on delivery, $0 dine-in and 10% on catering. That fee is never charged to the restaurant.
No. Direct orders route into the tablet you already run for the delivery apps, through Otter, an order aggregator, meaning software that pulls orders from several apps onto one screen. If your POS (point of sale, the register system) and KDS (kitchen display system, the screen on the line) already handle marketplace tickets, they handle these the same way.
Live in under 2 weeks, and typically a first direct order inside a week of signing up. Menu mapping, integration, QR codes on receipts and a short staff walkthrough are the whole setup.
Run the break-even formula with your real ticket. A $16 ticket at 28% saves $4.48 an order, so break-even is around 111 orders a month, roughly four a day. A lower ticket means more orders to break even, not failure. What kills the math is launching a direct site and never telling anyone.
Do not take our arithmetic on faith. Run yours. The free AI report at grader.zay-os.com does this math against your real volume and menu, then hands you the fix list. No sales call attached. Platform details at Zayos · live demo on a real customer · multi-location case study.
30 minutes. We look at your business and tell you exactly what we'd do. client or not.
Book Your Free Audit