Why Your Restaurant's Delivery Orders Are Unprofitable (And How to Fix It)

A 40-seat casual restaurant in Austin pulls in $180K monthly revenue. Looks solid until you break it down: 35% comes from DoorDash, Uber Eats, and Grubhub. But those platforms take 30% commission. The kitchen is running extra labor because delivery orders come in unpredictably. Some items don't travel well, but they're still getting ordered. After platform fees, labor overflow, and packaging, the owner is netting maybe 8-12% on delivery orders versus 28-32% on dine-in. This is the delivery trap most restaurants don't see.

The Hidden Cost Structure

Let's be specific. A $15 burrito ordered on DoorDash: platform takes $4.50 (30%), delivery bag and label costs $0.40, extra labor to pack and prep off-order-flow costs another $1.20 in fully-loaded labor, and condiments/napkins add $0.25. Your gross profit on a $15 order is now $8.65. Deduct those costs: $8.65 - $4.50 - $0.40 - $1.20 - $0.25 = $2.30 profit. You're making 15% on that order. Meanwhile, a dine-in $15 order costs you $4.50 in COGS, $0.50 in labor to plate and serve, and you keep $10. That's 67% margin.

We analyzed a 60-seat Thai restaurant in Denver tracking delivery orders for three months. They thought delivery was growing their business. Reality: 42% of delivery orders were items that didn't retain quality well (curries, soups). Customers complained. Refund rates on those items hit 11%, which triggered lower ratings and fewer order suggestions from the algorithm. Lower ratings meant DoorDash buried them, so they raised prices 12% to compensate for lower order volume, which triggered more complaints.

Three Levers That Actually Move Profit

• Optimize your delivery menu: Remove items with refund rates above 8%. You'll lose 5-8% of order volume but gain 18-22% profit per order on what remains.
• Raise delivery prices strategically, not universally: Keep your hero items (high-margin, travel-friendly) at platform parity. Raise others 8-12%. Most customers don't notice.
• Batch delivery orders for timing: If you control pickup timing (pickup location, time window), you reduce kitchen chaos. One pizza place reduced delivery prep time from 22 min to 14 min by batching orders.

Delivery isn't a profit channel. It's a customer acquisition channel. Price it like one—lower margins now, higher lifetime value later.

Building a Direct-to-Customer Escape Hatch

Every dollar you spend on DoorDash is a dollar that could build your own ordering channel. A family-owned sushi restaurant in Portland spent $800/month on Google Local Services ads promoting direct orders through their website. They built a simple ordering page using Square Online (free tier). After six months, 18% of their delivery orders shifted from platforms to direct. That 18% costs them $0 in commissions. On $4,200 monthly delivery volume, that's $756/month reclaimed.

The psychology works: frequent customers see the 'Order Direct' link in Google, your Google Business Profile, and email. They realize they save $2-3 per order. A taco shop we worked with offered a $3 discount for orders placed through their website instead of apps. It cost them margin, but they own the customer relationship and data. After one year, 31% of their delivery volume moved off platforms. They're now considering a small loyalty program—something platform apps won't let them do.

The Math on Going Platform-Light

• Keep platforms for new customer acquisition only: Don't bid for visibility. Let organic orders come in but deprioritize.
• Raise platform menu prices 10-15% to compress demand: Counterintuitive, but you want fewer high-friction orders on those channels.
• Invest the savings into direct channels: Google Local Services, email to past diners, retargeting ads pointing to your website.
• Measure CAC (customer acquisition cost) by channel: If platform CAC is $8 but direct CAC is $3, shift spend.

A Mediterranean bowl place in Seattle did the math three months ago. They cut Uber Eats and Grubhub to 15% of delivery volume (down from 65%) by raising prices there and promoting direct ordering. Their average order value on direct rose to $16 (from $14 on platforms). Delivery profit margin jumped from 12% to 31%. They didn't lose customers; they lost unprofitable orders.

← Back to all articles