Delivery App Markup - 83% or 15%? Why Both Numbers Are Right
It happened again last Friday.
I had a burger combo sitting in my cart, thumb hovering over "Place Order," and the total just... didn't match the number in my head.
Not by a couple of bucks. By a lot.
So I went looking for how much delivery apps actually mark things up. And I immediately hit a wall that I bet you've hit too.
One article says delivery costs 83% more. The next one says restaurants add about 15%. Same topic, same year, numbers that are five times apart.
Here's the thing: both are correct. They're just measuring two completely different things, and almost nobody says which one they're measuring.
That's what this post fixes. By the end you'll be able to look at any delivery app markup headline and know exactly what it's counting.
Why you keep seeing two completely different numbers
Short version: 15% is the menu. 83% is the entire bill.
The ~15% number describes only what happens to the price of the food itself. Restaurants list delivery menu prices roughly 10-20% higher than their in-restaurant equivalents, with the average landing near 15% (source: Sauce, a restaurant-technology publisher). Uber Eats itself "recommends a 10-15% price increase for restaurants to balance the commission cut," though plenty of restaurants push closer to 20% (source: iOrders/Nabe Eats analysis, 2026).
The 69-92% numbers describe the total you actually pay. The methodology behind them is explicit: the study combined "food, fees, taxes, as well as suggested tips" for every order (source: FinanceBuzz).
So when someone says DoorDash is 83% more expensive, they mean:
menu markup + delivery fee + service fee + sales tax + a suggested tip, all stacked on top of what you'd have paid standing at the counter.
Neither number is a lie. Only one of them is "the markup."
Where these numbers come from matters. The 92/83/80/69 platform figures are not government statistics. They come from FinanceBuzz, a personal-finance publisher that ran its own price-check study. The 100-city McDonald's data comes from Self Financial, a fintech company. Both disclosed their methods, and both methods are reasonable. But a lot of news outlets reported these numbers without naming the publishers, which is exactly why they circulate as if a federal agency produced them. They didn't.
How much more does DoorDash cost than in-store?
All-in, the most-cited study puts it at 83% - and DoorDash isn't even the worst.
Here's the platform breakdown from FinanceBuzz, in the version updated May 20, 2026:
| Platform | All-in markup vs. in-store |
|---|---|
| Postmates | 92% |
| DoorDash | 83% |
| GrubHub | 80% |
| Uber Eats | 69% |
The study priced a meal for one person - an entrée, a side, and a drink - from 10-11 popular chains, ordered on the same day within 45 minutes of each other, from locations within 3 miles of the delivery address, in a single city.
One wrinkle worth knowing, because you'll run into both versions in the wild: FinanceBuzz's own numbers changed. The January 2024 run reported Postmates up to 105%, DoorDash 95%, GrubHub 93%, Uber Eats 80%. Same publisher, same method, later run - lower numbers. Don't mix the two sets, and don't be surprised when an older article quotes 95%.
And a separate study backs up the general range. Self Financial priced an identical McDonald's order - a Big Mac Combo, a Quarter Pounder Deluxe Large Meal, and two Happy Meals - across the 100 largest U.S. cities, collected between 5-8 p.m. on weekdays as of April 1, 2025:
| How you order | Total | Markup |
|---|---|---|
| Directly from the restaurant | $36.95 | — |
| DoorDash | $63.21 | +71.1% |
| Uber Eats | $62.60 | +69.4% |
| GrubHub | $61.26 | +65.8% |
Sit with that middle number for a second.
You handed over an extra $26.26 on a $36.95 order. That's nearly enough to buy the entire family meal a second time - and you got exactly the same amount of food.
Location swings this hard, too. The most expensive city by total was Gilbert, Arizona at $90.61 (a 77.5% markup); the cheapest was Laredo, Texas at $45.39 (61.9%). The highest markup of all 100 cities was New Orleans at 237.8% above restaurant price. The lowest was Anchorage, Alaska at 42.8%.
Have you compared your own city's totals? The spread is bigger than most people assume.
Where does the 83% actually come from?
This is the part almost nobody shows you, and it's the whole ballgame. Once you see the fees stack, 83% stops sounding like a wild exaggeration and starts looking like plain arithmetic.
Let's walk a $25 order through it, using DoorDash's published consumer fee structure.
Step 0 - The food, in store: $25.00
Step 1 - Menu markup (10-20%, avg ~15%)
The restaurant lists the same items higher on the app. Your subtotal is already above $25 before a single fee appears.
Step 2 - Delivery fee: $0.99-$7.99
Varies by distance, demand, and the restaurant's partnership level.
Step 3 - Service fee: 10-15% of subtotal
This is a percentage, so it grows with your order.
Step 4 - Situational fees
- Small order fee: $2.00 on orders under $12
- Regulatory response fee: $1-$2 in cities with driver-benefit mandates
- Priority delivery (optional): $1.99-$3.99
Step 5 - Sales tax, applied to a bigger base than the in-store price.
Step 6 - Suggested tip (~15%), calculated on that inflated subtotal.
Add up steps 2 through 4 alone on a $25 order and you get roughly $8-$14 before tip - a 32% to 56% surcharge on the food cost by itself.
Now layer the 15% menu markup underneath and the 15% tip on top.
You land in the 70-90% range. That's the 83%.
Notice what happened in Step 1 and Step 3. The restaurant raised the menu price partly to cover the commission the app charges it. Then the app charges you a service fee on top of that raised price.
You pay for the commission twice. Once buried in the menu price, once again on your receipt.
Why is delivery more expensive than eating at the restaurant?
Because the restaurant is handing the platform 15-30% of every order, and that money has to come from somewhere.
The published commission rates:
- DoorDash offers Basic, Plus, and Premier partnership plans charging 15%, 25%, and 30% commission respectively, plus 6% on U.S. pickup orders that comply with in-store price matching and other terms.
- Uber Eats commissions run 15% to 30% depending on plan tier - a Lite Plan at 15%, a Plus Plan at 25%. Analysis of Uber's structure puts per-order charges anywhere from 7% to 30%, "with delivery commonly landing at 20-30%."
But the headline rate isn't what restaurants actually feel. Once payment processing fees, required promotions, and refunds get layered on, the effective cost commonly reaches 30-40% of the order total (source: Rezku, a restaurant POS vendor - note that POS vendors compete with third-party marketplaces, so read that estimate accordingly).
Think about a 30% effective take rate on a business that runs on thin margins. A restaurant has two choices: eat it, or price for it. Most price for it.
The blame loop nobody wins
Here's the part that keeps this problem stuck: neither side claims the markup.
DoorDash's position: restaurants independently set their delivery menu prices, and "DoorDash does not require restaurants to match in-store prices with delivery prices." DoorDash recommends pricing as close to in-store as possible, and caps markups at 10% for restaurants to qualify for "Most Loved," its recognition program. It also cites internal research across more than 4,500 partner restaurants finding that higher-markup restaurants saw 37% fewer sales and up to 78% lower reorder rates. DoorDash has publicly identified menu markups as one of the top customer complaints on its own platform, and has been testing an in-app label reading "Menu Matches In-Store Prices."
⚠️ Every DoorDash figure in that paragraph is company-published, internally conducted research from a party with an obvious interest in pointing the finger at restaurants. Useful, but not neutral.
Uber Eats' position: it doesn't mark up prices directly. It publishes a "Menu Markup" metric to merchants measuring how closely their delivery prices track in-store prices, and recommends keeping them consistent.
The restaurants' position: we're paying you 15-30%.
Both descriptions are accurate. Both parties are also describing the other one's fault.
And customers respond, whether or not they can name what's happening. When the in-app ticket runs 15% above in-store, people who used to order weekly slide to every 12-18 days. In higher-ticket categories ($17-$30), that price gap pushes cart abandonment up 18% to 27% (source: Santiago & Company, 2026 restaurant trends).
Which restaurant chains have the biggest delivery markup?
Chick-fil-A, by a wide margin - and the dollar example is genuinely startling.
From the January 2024 FinanceBuzz run, which published more chain-level detail:
| Chain | All-in markup range | Real example |
|---|---|---|
| Chick-fil-A | 82%-149% | $9.25 in store → $23.01 on GrubHub, $16.87 on Uber Eats |
| Starbucks & Dunkin' (tied) | 52%-102% | $9.48 in store → up to $19.12 |
| Cava & Jason's Deli (tied) | 47%-87% | — |
| Chipotle | 44%-77% | $18.42 in store → nearly $40 through Postmates |
| McDonald's | ~79% in the cited example | $11.18 → $19.96 on Postmates |
That Chick-fil-A line is worth translating. A $9.25 lunch became $23.01. You paid for two and a half of the same lunch and received one.
The current version of the study lists Chick-fil-A's range as 71%-134% - lower, but still the top of the board.
And here's the uncomfortable part: the "cheap" chains aren't cheap either. At the lower-markup restaurants like Chipotle, Jersey Mike's, and McDonald's, delivery still adds 44% to 77% to a meal.
One more number worth untangling. You may see Chick-fil-A's markup quoted as 20-30%. That figure comes from a class action complaint about Chick-fil-A's own app, and it refers to menu-price markup only - not the all-in total. Same company, three legitimate numbers, three different measuring sticks. This is the whole theme of this post.
Is the delivery markup actually shrinking?
Yes - but not for the reason you'd hope.
Self Financial ran its identical McDonald's order twice: once on September 25, 2023, once on April 1, 2025. Here's what changed:
- Ordering directly from McDonald's: $36.95 average in 2025, a 23.7% increase since 2023
- Ordering through third-party apps: $62.36 average, a 7.8% increase since 2023
- Average markup: 93.8% (2023) → 68.6% (2025)
Read those three lines together.
The markup fell almost 25 percentage points. But the delivery apps didn't get cheaper - they got 7.8% more expensive.
The gap closed because the restaurant's own prices shot up 23.7% and caught up to the delivery total.
So both of these headlines are true from the same dataset: "delivery markups are shrinking" and "delivery has never cost more." Whenever you see the first one, check whether the second one is doing the work.
Do delivery apps have to tell you prices are higher?
Almost nowhere in the U.S. And where disclosure exists, a lawsuit usually put it there - not a law.
As of August 2026, no general federal requirement forces a delivery platform to tell you that its menu prices differ from in-store prices.
The FTC's junk fee rule doesn't cover food delivery. The Rule on Unfair or Deceptive Fees took effect May 12, 2025, but it applies specifically to live-event ticketing and short-term lodging. Food delivery was never in scope.
That may be changing. On April 16, 2026, the FTC published an Advance Notice of Proposed Rulemaking (ANPRM) targeting potentially unfair or deceptive practices in fees charged by online food and grocery delivery platforms. The notice poses more than 60 questions, and the contemplated rule would address failures to disclose (a) the total price for delivery and (b) the "existence, nature, purpose, amount, refundability, or recipient" of any delivery fee.
Comments closed May 18, 2026, with 438 filed. The Commission now decides whether to move to an actual proposed rule. California AG Rob Bonta filed a letter supporting the rulemaking on May 19, 2026; the National Taxpayers Union argued the FTC "should exercise restraint before imposing sweeping regulations on food delivery platforms."
California already went further. SB 478, the state's honest-pricing law, took effect July 1, 2024 and requires advertised prices to include all mandatory fees except tax and shipping. It contains a restaurant exemption - but that exemption does not extend to food delivery platforms.
And New York City moved the other way. NYC once had a permanent 15% commission cap for delivery plus 5% for other fees, one of only two U.S. cities to make pandemic-era caps permanent. Then in 2025, following a June agreement with the platforms, the City Council approved a bill letting apps take up to 43% of each order: 15% for core delivery, 5% for basic marketing, 3% for card processing, and an additional 20% for enhanced services.
The one city that capped delivery commissions at 15% now allows up to 43%.
The disclosures that do exist came from courts
- Chick-fil-A - $4.4M settlement. Plaintiffs alleged the company promoted "free" or $2.99-$3.99 delivery while raising delivery menu prices 20-30% above in-store, effectively charging a hidden fee. It covered orders placed Nov. 1, 2019 through April 30, 2021 in California, Florida, Georgia, New Jersey, and New York; class members received $29.25. Chick-fil-A settled without admitting wrongdoing - and agreed to post a disclosure that prices may be higher for delivery.
- Grubhub - $25M FTC and Illinois AG settlement (Dec. 17, 2024). Regulators alleged Grubhub hid the true cost of its service with junk fees labeled "service fees" and "small order fees," listed restaurants without permission, and misled Grubhub+ subscribers about avoiding fees.
- DoorDash - $18M Chicago settlement, including allegations that it failed to disclose that prices on its app might differ from the restaurant's own menu.
- An ongoing price-fixing class action against Grubhub, Postmates, and Uber Eats over alleged "No Price Competition Clauses" that stop restaurants from charging delivery customers different prices than dine-in customers.
Read those last two together and enjoy the whiplash: one lawsuit attacks platforms for allowing price differences, another attacks them for contractually forbidding price differences.
The only standing disclosure I could find is a single sentence Uber added after pressure from state attorneys general: before you finalize an order, Uber shows "Prices may be lower in store." It appears in Pennsylvania and Washington, D.C. - and, per the reporting, only there. More recently, industry analysis notes that in 2025 Uber Eats and Rappi enabled explicit notices when in-app prices differ from in-store prices.
Notice what that sentence does and doesn't do. It tells you prices may be lower. It never tells you by how much.
Honestly, here's what these numbers can't tell you
I'd rather flag the limits than pretend this is settled science.
1. The headline markups are not official statistics. FinanceBuzz is a personal-finance publisher; Self Financial is a fintech company. Both self-published their studies, partly for traffic. Methods are disclosed and reasonable, but no government agency or peer-reviewed body produced these figures.
2. The FinanceBuzz sample is small and local. 10-11 chains, one meal for one person, within 3 miles of one delivery point, in a single city. Directionally useful; not a national census.
3. The FinanceBuzz numbers moved a lot between runs (105/95/93/80 in 2024 vs. 92/83/80/69 in 2026). Treat any single figure as a snapshot.
4. FinanceBuzz never states when its underlying prices were collected, even though the article carries a 2026 update date.
5. The 30-40% effective commission estimate comes from restaurant-technology vendors who compete with the delivery marketplaces. Published rates (15-30%) are verifiable; the effective rate is an estimate from an interested party.
6. Your order is not the study's order. Fees vary by distance, demand, city regulations, order size, and whether you tip. The genuinely official documents in this space are the FTC's April 2026 ANPRM, the FTC/Illinois v. Grubhub complaint and order, and California SB 478.
FAQ
Q. How much more does DoorDash cost than in-store?
A. All-in, FinanceBuzz measured 83% more as of its May 2026 update, counting food, fees, taxes, and a suggested tip. If you only compare the menu prices, the typical markup is 10-20%, averaging about 15%.
Q. Why is DoorDash more expensive than ordering directly from the restaurant?
A. Restaurants pay DoorDash 15-30% commission depending on partnership tier, and often raise delivery menu prices to offset it. On top of that raised price you pay a delivery fee ($0.99-$7.99), a service fee (10-15% of subtotal), sales tax, and typically a tip.
Q. Does DoorDash have to tell you prices are higher than in the restaurant?
A. There is no general federal requirement as of August 2026. Uber Eats displays "Prices may be lower in store" in Pennsylvania and Washington, D.C. only. Chick-fil-A agreed to post a delivery-price disclosure as part of a $4.4 million settlement. The FTC opened a food-delivery fee rulemaking on April 16, 2026, which is still pending.
Q. Which delivery app is cheapest?
A. In the FinanceBuzz study updated May 2026, Uber Eats had the lowest all-in markup at 69%, followed by GrubHub at 80%, DoorDash at 83%, and Postmates at 92%. In Self Financial's McDonald's study, GrubHub came out lowest at 65.8%.
Q. Does picking up the order myself actually save money?
A. It removes the delivery fee, service fee, and tip entirely. It does not necessarily remove the menu markup - DoorDash charges a reduced 6% commission on U.S. pickup orders that comply with in-store price matching, but whether a specific restaurant matches its in-store prices is up to that restaurant.
Q. Are delivery app fees getting cheaper?
A. Not in dollars. Between 2023 and 2025, McDonald's third-party delivery totals rose 7.8%. The percentage markup fell from 93.8% to 68.6%, but only because in-store prices rose 23.7% over the same period.
What to check before you hit "Place Order"
I'm not going to tell you to quit delivery apps. I'm not quitting them either.
Just do one thing tonight.
Before you check out, pull up that same item on the restaurant's own website or first-party app.
It takes about thirty seconds. And right now, that thirty seconds is the only reliable way to see the number nobody will show you.
Because here's what all of this comes down to. The delivery apps will tell you prices may be higher. The FTC is asking - as of April 2026, still asking - whether they should have to tell you more. Two jurisdictions get that one sentence of warning.
Not one of them tells you by how much.
How about you?
- I'll keep ordering - convenience wins 🛵
- I'm switching to pickup after reading this 🚗
- Apps should be required to show the in-store price 📱
Drop a 1, 2, or 3 in the comments - I'm genuinely curious where people land.
And if you want the next one of these, subscribe and I'll keep pulling receipts apart.
Comments
Post a Comment