Third-party delivery is killing your margin
Not because the apps are evil, but because they charge an acquisition price for customers they did not acquire. Here is the maths, and where the line actually is.
Written by GrowwithBA, a performance marketing agency that works with independent restaurants. Some links are affiliate links, disclosed below — they never change what we recommend.
Marketplaces take roughly 15–30% of every delivery order. That comes off the top — before food cost, before labour, before rent. A restaurant running a typical 3–5% net margin can hand its entire annual profit to commissions without ever seeing the line item. The apps are still worth it for discovery. They are terrible value for serving a regular who already knows your name.
The maths, on one plate of food
Take a $30 delivery order. The marketplace takes 25%, so $7.50 leaves before anything else happens. Food cost at roughly 30% is another $9. Labour, packaging, the driver's tip on your card statement, rent, utilities. What is left is measured in cents, and that is on a good order.
Now scale it. A restaurant doing $12,000 a month in delivery revenue at 25% hands over $36,000 a year. On a 3–5% net margin, a million-dollar restaurant keeps $30,000–$50,000. The commission line can be larger than the profit line. That is not a marketing problem. It is arithmetic.
Most owners cannot name what fees cost them per year, and you cannot manage what you have not measured. Run your delivery revenue through the fee calculator. It takes ten seconds, and it will tell you honestly if switching is a bad idea at your volume.
When marketplaces are genuinely worth it
This is where most "escape the apps" advice goes wrong. Marketplaces are a customer acquisition channel, and judged as one they are reasonable. A 25% commission to reach a diner who has never heard of you is a marketing cost, not a robbery. Compare it to what you would pay for that same first order through paid ads.
The damage is done on the second order, and the twelfth. Paying 25% to serve a regular who already knows your name, your menu and your address is where the margin dies. The apps charge the same acquisition price for a customer they did not acquire.
A first-time diner discovering you. A slow Tuesday you would not otherwise fill. A new neighbourhood you are testing.
A regular ordering their usual. A customer who searched your restaurant by name. Anyone who would have found you anyway.
How to stop the bleed without losing the reach
- Measure it. Put your real delivery revenue into the calculator. Know the annual number before you touch anything.
- Keep the apps for discovery. Do not delete your marketplace listings. They are how new diners find you.
- Give regulars somewhere better to order. Your own ordering page, with the same menu and no inflated prices.
- Point every free channel at it. Google Business Profile, Instagram bio, your website, the flyer in the bag. All should link to your ordering page, not a marketplace.
- Put a card in every marketplace order. "Order direct next time and skip the app fees." It costs pennies and it converts the customer the app charged you 25% to reach.
The goal is not zero marketplace orders. It is that your repeat orders happen somewhere you keep the margin and the customer.
A flat fee only wins above a certain volume
Owner.com gives restaurants a branded ordering website and app with 0% restaurant commission on its Flat plan, and the customer's contact details stay with you. Restaurants it has published: Cyclo Noodles saved $31,000 in third-party fees, Doo-Dah Diner $19,000, Talkin Tacos grew direct sales 970%.
But a flat monthly fee is only cheaper than a percentage above a certain delivery volume. Below it, commissions genuinely cost you less and switching would lose you money. We would rather you saw that in the numbers than found out after signing up — run your own figures.
Figures published by Owner.com. Results vary by restaurant, volume and market.
Commission is the cost you can count. Alongside it sit the customer whose contact details the marketplace keeps, the Google listing that sends the order to an app, and the one-time diner nothing brought back.
Read the five margin leaks, in the order to fix them →Want us to find your leak for you?
Send your restaurant's website. We will audit your ordering setup, Google Business Profile, local SEO and review flow, then email a prioritized fix list. Free, 48-hour turnaround. If switching platforms is wrong for your volume, we will say so.
Frequently asked
How much do third-party delivery apps take from restaurants?
Roughly 15–30% of each delivery order depending on your plan, plus payment processing. DoorDash, Uber Eats and Grubhub sit in a similar range, and rates are set nationally rather than by city.
Is third-party delivery ever worth the commission?
Yes — as a customer acquisition channel. Paying a commission to reach a diner who has never heard of you is a marketing cost, and often a reasonable one compared with paid ads. The damage is paying that same commission on the second, fifth and twentieth order from a regular.
Should I remove my restaurant from DoorDash?
Usually not. Removing the listing loses you discovery. The better move is to keep marketplaces for new customers and give your regulars a direct ordering channel with the same menu and no inflated prices.
Will a flat-fee ordering platform save me money?
Only above a certain delivery volume. Below that point a flat monthly fee costs more than commissions and switching would lose you money. Put your real revenue into a fee calculator before changing anything.
What is the fastest way to reduce delivery commissions?
Put a card in every marketplace order inviting the customer to order direct next time, and point your Google Business Profile, Instagram bio and website at your own ordering page rather than a marketplace. Both cost almost nothing.
Related: why your restaurant is busy but not profitable · why you have no repeat customers · DoorDash fee calculator · how to increase direct orders
Affiliate disclosure: Owner.com is an affiliate partner. If you sign up through our link we may earn a commission, at no extra cost to you. It does not change the advice above, and we say plainly when a tool is the wrong fit.