The short answer
Phone orders are dramatically more profitable than delivery app orders. Third party platforms take 15% to 30% of every order per their published plan tiers, while a phone order costs only processing fees and the labor to answer. The catch was always missed calls, and that is the problem AI answering removes.
What Delivery Apps Actually Take
Marketplace commissions on the major platforms are published tiers: DoorDash's Basic, Plus, and Premier plans run 15%, 25%, and 30% of the order subtotal, and Uber Eats' delivery plans span the same 15% to 30%, before payment processing and optional marketing fees are added on top. On a $40 order at a 30% tier, $12 leaves the building before food cost, labor, or rent is paid. The platform also keeps the customer relationship and the data, so the restaurant cannot remarket to its own buyer or even see who ordered.
Per industry benchmarks published by Restaurant365 and Toast, independent restaurants operate on net margins of 3% to 5% per benchmarks published by Restaurant365 and Toast. A commission that consumes 15% to 30% of the ticket does not trim the profit on that order. It erases it several times over, which is why many operators treat app volume as paid advertising rather than as revenue.
The Margin Math on a Typical Order
A phone order is the highest margin ordering channel a restaurant owns. The costs are card processing, typically under 3% per published 2026 fee guides, and the minute or two of labor to take the call. There is no commission, no per order marketing fee, and no intermediary between the restaurant and the customer's phone number. The customer relationship stays in the building, which is worth more over time than any single ticket.
Run the comparison on that same $40 ticket. On the phone, the restaurant nets roughly $38 to $39 after processing. Through a 30% commission app, it nets about $28 before any of its own costs. Across 300 orders a month, the gap is on the order of $3,000, every month, for identical food going out the same door.
Why Restaurants Lose Phone Orders Anyway
The phone channel has one fatal weakness: someone has to answer it. Industry analyses suggest restaurants miss a meaningful share of incoming calls, with peak hour miss rates commonly estimated at 20% to 30% or worse. A caller who hits voicemail does not wait. They open a delivery app, and the restaurant pays commission to win back its own customer.
Every missed call is a double loss. The restaurant forfeits the high margin sale and often buys the low margin version of it back through a platform, paying commission for a customer it already had. The channel that should be the most profitable ends up neglected precisely because it depends on the busiest people in the building answering during the busiest hours of the week.
Making the Phone Channel Reliable
The fix is not asking a slammed kitchen to answer faster. It is removing the dependency on a human picking up. When every call is answered instantly, the order is taken completely, and payment is collected on the line, the phone competes with the apps on convenience while keeping the full margin in the restaurant.
X1 Voice from Batch Group answers a restaurant's existing phone line 24/7, takes complete orders with modifiers, collects payment on the call, and sends the ticket directly into the POS.
How Do the Channels Compare Order by Order?
The published economics, side by side.
| Phone / direct order | Delivery app order | |
|---|---|---|
| Platform cut | None | 15% to 30% of subtotal on DoorDash's published Basic, Plus, and Premier tiers; Uber Eats spans the same range |
| Payment cost | Card processing, typically under 3% per published 2026 fee guides | Included in or added to commission |
| Customer data | Yours: number, history, preferences | The platform's |
| Discovery | None: the customer already chose you | The platform's marketplace audience |
| Labor | Answering and keying the call, unless automated | None at order time |
| Best fit | Regulars, large orders, catering | New-customer acquisition |
What Does a Month of Volume Look Like Shifted?
A shop doing 600 delivery-app orders a month at a $30 average, $18,000 of subtotal volume, on a 25% published tier pays about $4,500 in commission.
Shift a third of those orders, the regulars who already know the restaurant, to the phone or direct site. That $6,000 of volume now costs roughly $180 in processing at a sub-3% published rate instead of $1,500 in commission: about $1,320 a month recovered, roughly $15,800 a year, from customers the platform did not discover for you.
Per industry benchmarks published by Restaurant365 and Toast, independent restaurants run 3% to 5% net margins, so $15,800 of recovered commission on this illustrative shop rivals the entire annual profit of a quarter million dollars in top-line sales. That is why channel shift, not menu-price surgery, is the highest-leverage fix on the P&L.
When Are the Delivery Apps the Right Choice?
The apps earn their commission in real cases, and the honest strategy uses them for what they are. A new restaurant with no audience buys discovery: the marketplace puts it in front of thousands of nearby customers no ad budget could reach as efficiently. Ghost kitchens and delivery-only brands live entirely on that discovery. Slow dayparts can be profitably filled at a 15% Basic tier even when a 30% tier order is marginal. The failure mode is not being on the apps; it is letting the apps keep the regulars. Acquire on the marketplace, then move repeat orders to channels you own.
How Do You Shift Orders Off the Apps?
The playbook, in the order that works.
- Answer every phone call: a missed call is an order pushed back to the app, and AI answering removes the miss
- Box-topper and receipt inserts with the direct number and a first-order incentive
- Price transparently: many operators list slightly higher on the apps to reflect the published commission
- Capture direct-order customer numbers and market to them yourself
- Keep the apps for discovery at the tier whose math works, per their published plans
Commonly Asked Questions
- How much do delivery apps charge restaurants?
- Marketplace commissions generally range from 15% to 30% of the order subtotal, with additional fees for payment processing and promoted placement.
- Should restaurants leave delivery apps entirely?
- Not necessarily. Apps can work as customer acquisition. The mistake is letting commissioned orders replace direct channels like the phone, which carry no commission at all.
- Do customers still order by phone?
- Yes. Phone ordering remains a major channel for pizza, takeout, and catering, and it skews toward larger tickets. The constraint is usually the restaurant answering, not the customer calling.
- How much do delivery apps really take per order?
- DoorDash's published plans take 15%, 25%, or 30% of the subtotal by tier, and Uber Eats' delivery plans span the same range. On a $30 order at the middle tier that is $7.50 before your food, labor, or rent, against roughly ninety cents of processing on a direct order.
- Should a new restaurant skip the apps entirely?
- Usually no. The marketplaces are the cheapest discovery a no-audience restaurant can buy, per their reach at the published 15% entry tier. The discipline is converting the second order to a direct channel rather than renting your regulars forever.
- What is the fastest way to recover commission dollars?
- Stop missing phone orders. Every answered call is an order at processing-fee economics instead of published commission tiers, which is why pairing the phone line with AI answering is typically the highest-ROI first step, ahead of loyalty programs or price changes.
Keep reading