How Do Delivery Apps Hurt Restaurants? The Full Cost Breakdown

Delivery apps hurt restaurants in one specific, measurable way: they take a percentage of the order value, while restaurants earn profit on the margin.

A typical restaurant runs on a pre-tax profit margin of roughly 5%, and a 15–30% commission is applied to the whole ticket.

On a $40 order, the platform takes around $12, while your own pre-tax profit on that ticket is closer to $2. Volume goes up, the payout doesn't — and that gap appears before marketing fees, refunds, and the customer record you never receive. Here is where the money and the control go.

1. Commissions are charged on revenue, not on profit

Commissions run from 15% to 30% of the order value, depending on the plan you sign — see how much Uber Eats charges restaurants for a worked example.

What matters is not the commission itself but the commission set against what the order was ever going to earn.

A typical restaurant works on a pre-tax margin of about 5%, and 42% of U.S. operators reported no profit at all in 2025 (NRA 2026 State of the Restaurant Industry). A 30% commission removes roughly six times the profit that ticket would have produced.

Here is the same $40 order run through both channels:

The point is not that $2.80 is a catastrophe on one ticket. It is that you carry every risk — food cost, labor, packaging, the complaint — for a fifth of the contribution, and whether that works depends on your average restaurant profit margin. Over ten years on its own channel, Sushi Kushi saved $1.5M in commissions it would otherwise have paid at 15%.

2. The commission is only the first line of the bill

Ask an operator what delivery costs, and you will hear the commission number. The payout statement tells a longer story.

Promoted listings, payment processing, discounts you opted into, and refunds issued to customers are all deducted before the transfer lands — which is why the delivery fee a customer sees, and the deduction you absorb, are different figures.

This is documented in one place. New York City is the only U.S. city with a permanent cap on what delivery platforms may charge restaurants, and in 2025 that cap was rewritten:

Read that as a ceiling, not an average — and note it exists because it had to be legislated. Outside New York, there is no cap, so you pay whatever plan you signed. Check DoorDash's fee tiers and what Grubhub costs a restaurant against your last statement, not the rate you remember agreeing to.

The 20% line is what buys position in the feed. Placement is bought, not earned: pay and you rise, stop paying and you drop, often within days.

3. Raising your app prices to cover the fees costs you orders

The standard advice is to raise menu prices on the apps to cover the commission. The platforms' own merchant data argues against it.

• DoorDash's merchant guidance, based on an internal study of more than 4,500 restaurants, reports that restaurants which mark up their app prices can see up to 37% fewer sales and up to 78% lower reorder rates.
• Uber's merchant documentation puts customer tolerance at around a 10% markup and states that when users notice a markup, only about 22% complete the order.

So marking up to cover a 30% commission does not neutralize it. It turns a margin problem into a volume problem, and the second is harder to see on a statement. You pay the commission anyway, on fewer orders, from customers less likely to return.

4. You don't get the customer — the platform does

Every marketplace order creates a customer record. It just isn't yours. You get the order and the payout; the platform keeps the name, email, phone number, and order history. You cannot email that person a Tuesday offer, enroll them in a restaurant loyalty program, or tell whether tonight's customer has ordered forty times before or once.

That is also why exposure rarely turns into loyalty: someone opening a delivery app is comparing delivery times and discounts, not choosing your restaurant.

The Wind-Chill Factory, a single-location restaurant in Ticonderoga, NY, runs its own channel and holds 4,234 customers it can actually contact — which is why 52% of its orders now come through its own app. For that record to be usable rather than just stored, it has to sit somewhere you own the customer data.

5. You absorb the reputation damage for a delivery you didn't control

Your kitchen's responsibility ends when the bag leaves the counter. The customer's judgment does not. A late driver, a cold bag, or a wrong address becomes a one-star review with your name on it, and you find out after the fact, with no way to contact the person and put it right.

Refunds work the same way. Platforms often settle complaints by refunding the customer and deducting it from your payout, sometimes for orders your kitchen got right. You pay for the failure, and you inherit the review, and both follow you onto restaurant review sites where your next customers are deciding.

6. Every extra platform adds a tablet, not a system

A food delivery app arrives with its own tablet, interface, and menu to update separately.

During a Friday rush, that means staff watching three screens, re-typing tickets into the POS by hand, and finding that an item you 86'd two hours ago is still selling on one platform.

Those errors come back as refunds and reviews, so the operational cost becomes a financial one. The fix isn't fewer orders — it's pulling every channel into one order screen.

7. The apps change the market you compete in, not just your margin

The damage is not only per-order. Research published in the Strategic Management Journal, covering U.S. restaurants from 2012 to 2018, found that when third-party delivery services enter a market, restaurant exits rise and the industry becomes more concentrated, with new openings failing to offset the closures (Raj & Eggers, 2026).

The effect is uneven: younger, smaller, and less efficient independents are most exposed because delivery removes proximity as a reason to choose one restaurant over another.

A customer who once chose the nearest Thai place can now choose any of them. Being on the apps doesn't protect you from that—it is simply the market you now compete in.

Where delivery apps still earn their keep

Marketplaces are good at one thing: introducing you to people who have never heard of you. For a new restaurant, a new neighborhood, or a quiet weekday shift, that is worth paying for.

The problem starts when discovery becomes permanent — when a regular's fortieth order still arrives through someone else's app. If the last mile is what stops you from changing that, you can dispatch deliveries without your own fleet.

How to cut the damage

You do not have to leave the platforms to stop losing money on them. Three moves, in order of return:

1. Price the channel, not the plate. Work out what each platform costs you per order, then decide whether a given item belongs there at all. Some dishes are not worth selling at a 30% commission at any price.

2. Give people a reason to order direct. A 10% discount is cheaper than a 30% commission. Put the offer in the delivery bag, on the receipt, and on stickers — aimed at customers who have already ordered twice, because those are the ones a marketplace charges you for repeatedly.

3. Own the channel. A commission-free online ordering system on your own site turns every repeat order into margin instead of a fee, and you can build your own restaurant website without a developer. More on how to increase your delivery sales once it is live.