The Real Math on Third-Party Delivery and Ordering Fees

What the Apps Actually Take From Every Order

Pull up your last DoorDash or Uber Eats statement. Not the dashboard summary. The actual payout report.

Most restaurant owners we talk to in Sarasota have never done this line by line. When they do, the number is usually worse than they thought. Marketplace delivery commissions typically run 15% on the cheapest tier, 25% on the mid tier, and up to 30% on the plans that promise better placement in the app. Pickup orders through the same apps still cost roughly 6%.

On a $40 order at 25%, the app keeps $10. You cover the food cost, the labor, the packaging, and the rent. The app covers an algorithm.

Run that across a month and it's often the biggest line item you've never budgeted for. A restaurant doing 400 marketplace orders a month at a $35 average ticket is sending roughly $3,500 a month to the apps. That's $42,000 a year. For a lot of independent restaurants, that's more than the profit.

Why Owners Tolerate It (And Why That Logic Half Works)

The standard defense: "those are orders I wouldn't have gotten anyway."

Sometimes that's true. A tourist standing on Siesta Key scrolling DoorDash has never heard of you. The app put you in front of them. That order is genuinely incremental, and paying 25% for a brand-new customer isn't crazy. It's a marketing cost.

But here's where the logic breaks: a huge share of app orders come from people who already know you. Your regulars. The family that's ordered from you eleven times. They search your name in the app because it's the path of least resistance, and you pay 25% on a customer you already earned.

Paying an acquisition fee once makes sense. Paying it on order eleven doesn't.

The Worked Example: When Direct Ordering Pays for Its Build

Let's do the actual math instead of hand-waving. Say you're doing 400 app orders a month at a $35 average ticket, on a 25% commission plan.

Scenario	Monthly orders	Fees paid	You keep
All orders through apps (25%)	400	$3,500	$10,500
40% shifted to direct (~3.5% processing)	160 direct / 240 app	$2,296	$11,704

Direct ordering on your own website isn't free. You'll pay card processing (roughly 2.9% plus 30 cents) and possibly a flat per-order fee to your ordering system. Call it 3.5% all-in. But 3.5% versus 25% is a 21.5-point swing on every order you move.

Shift just 40% of that volume to your own site and you keep about $1,200 more per month. A Foundation Build with ordering wired in starts at $2,500. At that pace it pays for itself in roughly two months, and everything after that is margin you were donating to a company in San Francisco.

Your numbers will differ. Run yours with your real ticket average and volume before you decide anything. If the savings come out under $300 a month, the build can wait. If they're over $1,000, you're funding someone else's growth.

When the Marketplaces Still Make Sense

We're not going to tell you to delete the apps. That advice sounds bold and costs restaurants real money.

• Discovery. New restaurants, new locations, and anyone chasing the snowbird wave from January through April benefit from being findable by people who don't know they exist yet.
• Delivery logistics. If you don't have drivers, the apps solve a genuinely hard problem. Running your own delivery fleet is a business unto itself.
• Slow-night volume. A 25% commission on a Tuesday order that fills otherwise-idle kitchen capacity can still beat an empty ticket rail.

The honest framing: marketplaces are a customer acquisition channel, not an ordering system. Use them to meet people. Then give those people a cheaper way to come back.

How to Actually Move Orders to Your Own Site

Building direct ordering and hoping people find it doesn't work. You have to route them there on purpose.

1. Make direct ordering stupidly easy. If a hungry customer can't get from your homepage to a confirmed order in three taps, they'll bail back to the app. We wrote up exactly how that flow should work in the three-tap rule for restaurant websites.
2. Fix your Google Business Profile ordering link. Google often defaults your "Order Online" button to a third-party app. You can change it to point at your own site. Most owners never check.
3. Put an insert in every app order. A card in the bag: "Order direct at our site next time, use code DIRECT for 10% off." You're spending 10% to stop spending 25%. Check your marketplace agreement first; most allow bag inserts, some restrict what they can say.
4. Answer the phone, or catch it when you can't. Plenty of orders still start as calls, and a missed call during the dinner rush is a lost ticket. Our free missed-call revenue calculator puts a dollar figure on how many you're dropping.
5. Price the apps honestly. Most marketplaces let you run higher menu prices in-app than on your own site. Customers who compare will learn that direct is cheaper. That's the point.

None of this is exotic. It's plumbing. We've watched Sarasota restaurants pick up real volume just by fixing the ordering path and the Google listing, before touching ads or anything fancy.

The Decision, In One Paragraph

Keep the apps for what they're good at: putting you in front of strangers. Build direct ordering for what it's good at: keeping the margin on people who already love you. The break-even math above tells you when the build is worth it, and for most restaurants doing more than a couple hundred app orders a month, it is. If you want a second set of eyes on your numbers, our restaurant work page shows how we approach these builds, and the math conversation costs nothing.

Marketplaces are rent. Your own ordering page is equity.

Frequently Asked Questions

How much do third-party delivery apps charge restaurants?

Marketplace delivery commissions typically run 15% on basic plans, 25% on mid tiers, and up to 30% on plans with better in-app placement. Pickup orders through the same apps usually cost around 6%. On a $40 delivery order at 25%, the app keeps $10 before you cover food, labor, and packaging.

Should restaurants leave DoorDash and Uber Eats entirely?

Usually no. The apps are genuinely good at discovery, reaching tourists and new customers who don't know you, and at delivery logistics if you have no drivers. The smart play is keeping them as an acquisition channel while routing repeat customers to direct ordering on your own website, where fees drop from 25% to roughly 3.5%.

When does direct online ordering pay for its build cost?

Run the math on your own volume. A restaurant doing 400 app orders a month at a $35 average ticket that shifts 40% of orders to direct keeps about $1,200 more per month, so a $2,500 website build with ordering pays back in roughly two months. Under a couple hundred app orders a month, the savings shrink and the build can wait.