The Real Cost of "Free" Delivery Platforms
Javier Trujillo runs five restaurants in Omaha, Nebraska: Javi's Tacos, Time to Rise and Shine, Frosty Mug, Helados Locos, and El Milagro. Business was good. Orders were up. The tablets kept pinging. Then he pulled his annual numbers.
In 2025, Trujillo paid $188,000 in commissions to delivery platforms — spread across DoorDash, Uber Eats, and Grubhub. That figure didn't include the food cost on orders that were prepared but never picked up because no driver was available. As he told Omaha's First Alert 6 in April 2026, those abandoned orders were pure loss. He dropped every delivery app shortly after.
Trujillo's story, reported by Entrepreneur in April 2026, is not unusual. What makes it notable is the number attached to it. Most independent restaurant owners never calculate the aggregate. They see the per-order commission — 25%, maybe 30% — absorb it as a cost of doing business, and move on. The aggregate is where the picture gets ugly.
The headline rate is a starting point, not a total
The commission structures are public. DoorDash runs three tiers: Basic at 15%, Plus at 25%, and Premier at 30%. Higher tiers buy better placement in search results and eligibility for DashPass, the platform's subscription program. Uber Eats operates a similar ladder, ranging from 15% to 30% depending on plan level and market. In March 2026, Uber raised rates across most tiers for the first time in nearly a decade. Grubhub's base rates run 5–25%, but promoted placement and marketing add-ons push effective costs to 30% or higher.
Those headline percentages, though, are the beginning of the accounting, not the end. An August 2025 analysis by ActiveMenus, drawing on data from McKinsey and the Digital Restaurant Association, found that the actual cost of third-party delivery can exceed 40% of order revenue once hidden charges are included. Payment processing adds 2.9–3.5% per transaction — at rates typically higher than what restaurants pay through their own processors. Platform-side marketing and promotional fees tack on another 1–5% for boosted visibility. And refund chargebacks for customer complaints are deducted from the restaurant's payout, even when the issue was the driver's.
The math gets worse when you work backwards from it. The average independent restaurant operates on a net profit margin of 3–5%, according to data cited consistently by the National Restaurant Association, Toast, and NYU's restaurant finance analyses. A 25–30% commission doesn't just eat into profit. It eliminates it entirely and keeps going.
To avoid losing money on every delivery order, restaurants mark up their platform menus. The required markup isn't intuitive. A 20% commission requires a 25% price increase to break even. A 25% commission requires 33%. At a 30% commission, the break-even markup is 42.9%. One restaurant owner cited in a Daily Bruin report described a burrito that costs $7–8 in-store selling for $21 on a delivery app — with the restaurant receiving only $4 of that total after fees.
This is the engine: restaurants inflate prices to survive the commission, customers see inflated prices and associate them with the restaurant's brand, and the platform takes its cut from the inflated number. The restaurant's margin stays razor-thin. The platform's revenue grows.
How much revenue? DoorDash reported $13.7 billion in 2025 sales, a 28% year-over-year increase, and posted its first annual profit: $935 million. Over 3.1 billion orders were processed. The company now holds roughly 56% of the U.S. food delivery market, with Uber Eats at 23% and Grubhub at 16%, according to Statista and Business of Apps. Americans placed 903 million orders through DoorDash alone last year, per the company's Q4 2025 earnings release.
The platforms built a profitable business. The question is whether the restaurants powering it did.
The data they keep is worth more than the commission they charge
The commission conversation dominates most coverage of delivery economics, but several restaurant operators and industry analysts argue that the larger cost is invisible on any invoice: customer data.
When someone orders through DoorDash, the platform captures their name, email, phone number, delivery address, order history, and spending patterns. The restaurant gets the order and the payout. It does not get the customer.
DoorDash has been explicit about this. When New York City Council Member Keith Powers introduced a bill in April 2026 requiring delivery platforms to share customer data with restaurants at the time of order, DoorDash opposed it, citing consumer privacy concerns. A company spokesperson said DoorDash "cannot support efforts to force delivery platforms to share customer data without customer consent." Grubhub, facing the same bill, pointed to its Grubhub Direct product — a tool that lets restaurants build their own ordering pages and collect data from those transactions, separate from the marketplace.
The practical effect is that restaurants on delivery platforms cannot build email lists, run targeted promotions, or identify their most valuable repeat customers from platform orders. A PYMNTS analysis from March 2026 described the dynamic bluntly: restaurants have been repositioned from independent brands with direct customer relationships to suppliers within a marketplace. The platform controls visibility, pricing dynamics, and — crucially — data.
Paytronix, which provides loyalty and ordering technology to over 1,800 restaurant brands, published data quantifying the gap. About 41% of orders placed through first-party channels (restaurant websites and apps) come from loyalty program members. For third-party marketplace orders, that figure is 3%. First-party customers tip an average of $4.44 per order versus $1.89 on third-party platforms. Their average item price is $12.35 compared to $10.65. And they come back sooner — every 33 days versus every 43 days.
ActiveMenus framed this as "The Anonymous Customer Problem." When someone orders via DoorDash, they tell friends they ordered Thai food last night, not that they ordered from Bangkok Garden Restaurant. The platform gets credit for the experience. The restaurant becomes interchangeable.
Meanwhile, the restaurant's own order data trains the platform's recommendation algorithms — which surface competitors alongside the restaurant's own listing. DashPass and Uber One, the platforms' subscription loyalty programs, compound this: customers earn rewards from the platform, not the restaurant. The incentive structure points away from brand loyalty and toward platform loyalty.
What happens when you actually quit
Dave Query has been in the restaurant business in Colorado since 1994. His Big Red F Restaurant Group operates 12 locations across Boulder, Denver, Fort Collins, and Colorado Springs — including Jax Fish House, Centro Mexican Kitchen, and The Post Chicken & Beer.
Big Red F was an early DoorDash partner, signing up in 2017. During the pandemic, delivery orders from DoorDash, Uber Eats, and Grubhub accounted for as much as 40% of sales at some locations on busy nights. When dine-in returned, delivery kept growing. In early 2025, the group decided to accelerate — investing in advertising on the platforms, with the apps matching the spend.
The results forced a reckoning. As Query told Restaurant Business Online in January 2026, the revenue looked strong, but the margins were unsustainable at 30% commission rates. He pulled all Big Red F restaurants off the delivery marketplaces in September 2025.
The outcome was instructive and honest: total revenue went down. Margins went up. Query was comfortable with the trade-off.
Big Red F didn't abandon delivery. It moved to a direct model — customers order through each restaurant's own website, and DoorDash still handles the actual delivery fulfillment behind the scenes. The difference: a flat $4.99 delivery fee regardless of order size, and the restaurant keeps the customer relationship. As Query told the Boulder Reporting Lab, the arrangement is better for everyone — the customer pays less, and the restaurant stops hemorrhaging margin on every order.
Big Red F isn't alone in the shift. A Market Leader report cited by Restaurant Business Online found that 53% of operators said they are actively trying to reduce their reliance on third-party delivery. A separate survey by tech provider Qu found that 40% of limited-service restaurant brands believe first-party digital ordering will drive the most revenue growth in 2026 — far ahead of third-party delivery at 7%. Loyalty technology provider Thanx reported that nearly two-thirds of its restaurant clients are seeing first-party ordering grow faster than third-party as a share of digital sales.
The counter-argument: when the platforms actually earn their cut
None of this means delivery platforms are worthless. For certain restaurants in certain situations, the economics do work — or at least break even in exchange for something valuable.
A peer-reviewed study published in Management Science examined the impact of DoorDash, Grubhub, and Uber Eats on restaurant demand in the Chicago market. The researchers found that delivery platforms increased restaurants' total takeout sales, with approximately 3.8% of platform orders being genuinely incremental — new demand that would not have existed otherwise. More notably, being listed on delivery platforms increased dine-in visits by 6.2%, likely through an advertising and awareness effect. Customers discovered restaurants on the app, then showed up in person.
The catch: the study found that the benefits were substantially uneven. Chain restaurants saw incremental takeout growth roughly three times higher than independents. The platforms' algorithmic recommendation engines, brand recognition dynamics, and search placement all favor larger operators with more data and higher volume.
DoorDash's own research supports the cross-channel argument. According to the company's 2026 Restaurant Reservations and Operations Trends Report, 74% of consumers who dine in at a restaurant later order delivery from the same place, and 62% of delivery customers later dine in. This cross-pollination is real. The question is whether the restaurant can afford to pay 25–30% commission on the delivery side to generate the dine-in traffic.
For new restaurants or new locations building initial awareness, the answer may be yes. A restaurant that doesn't exist in a customer's mental map needs discovery, and delivery platforms are where a lot of discovery now happens — 51% of consumers turn to a third-party platform when deciding where to order, per DoorDash. Moe Shrikian, executive vice-president of the Hungry Howie's pizza chain, described platforms as "one tool in the toolbox" to Restaurant Dive in April 2026.
And from the customer's perspective, the experience gap between first-party and third-party has nearly closed. A 2025 study by Intouch Insight, reported in Restaurant Business Online, found that customer satisfaction was 88% for first-party delivery orders and 90% for third-party. The average price difference was just 44 cents — less than 2%.
The platforms are not irrational actors. They provide logistics infrastructure, a massive audience, and on-demand driver networks that most independent restaurants cannot replicate alone. The problem is not that the service has no value. The problem is that the pricing model extracts more value than most independent restaurants can sustain.
The question isn't whether to deliver
The National Restaurant Association's 2025 report found that 47% of U.S. adults order takeout at least once a week. More than a third order delivery weekly. Consumer demand for off-premises food is not declining. It is structural.
The real question is who owns the transaction. When a customer orders through a restaurant's own channel, the restaurant keeps the margin, keeps the data, and keeps the relationship. Paytronix's 2024 Online Ordering Report found that guests who order directly from a brand order 35% more items per check than those ordering through marketplaces. Their lifetime value is 35% higher than in-store-only customers. First-party mobile app users show 45% higher customer lifetime value than web users.
These are not small differences. They compound over every order, every week, every year.
The 2026 restaurant landscape, as the NRA's chief economist Dr. Chad Moutray put it, will require operators to "get the math right in a still-challenging economic environment." Sixty percent of operators reported softer customer traffic in 2025. Margins are thin. Costs are rising. In that environment, paying 25–30% of every delivery order to a platform that keeps your customers is not a growth strategy. It's a subsidy — from the restaurants that make the food to the platforms that move it.
Javier Trujillo did the math. Dave Query did the math. The $188,000 question facing every independent restaurant owner is whether they will, too.