DoorDash Commissions vs Direct Ordering: The Real Math

You are not arguing about an app. You are arguing about the only margin you have.

The National Restaurant Association's 2025 Restaurant Operations Data Abstract, built on financial data from more than 900 restaurants, found that full-service restaurants reported a median income before taxes of 2.8% of sales, and limited-service restaurants 4.0%. That is the whole business, after everything.

Now put a 25% or 30% commission on top of an order in that business. The order does not become slightly less profitable. It becomes a different economic event — one you agreed to because the app brought you the customer. The question is whether it should keep bringing you the same customer, month after month, at the same rate.

This is a website problem wearing a delivery-app costume. Below is the arithmetic, from published rates only, with the honest verdict at the end.

What does a DoorDash order actually leave on your bottom line?

On a $40 DoorDash delivery order at the 25% Plus rate, DoorDash keeps $10 and deposits $30 — before you have paid for a single ingredient. DoorDash's Marketplace pricing page lists three plans: Basic at 15% commission per delivery order, Plus at 25%, and Premier at 30%. Pickup is 6% for partners whose DoorDash pickup prices match their in-store prices.

Credit card processing is bundled into that commission — DoorDash states that all its Marketplace offerings include it. So the headline rate is closer to the true rate here than most people assume. The rest of the cost stack is small: no activation, subscription, software, or cancellation fee, and a $6-per-week tablet fee in the US after the free trial if you use their hardware.

The number that matters is not the commission. It is what is left after commission, food, and packaging — your contribution margin per order. That is the only figure that tells you whether an order is worth cooking.

How do the commission tiers differ, and which one are you on?

Every major marketplace runs the same three-tier structure, and the top tier is 30% on all of them. Here are the published US rates, all confirmed on the platforms' own pricing pages as of July 2026.

The rates come from DoorDash's Marketplace page, Uber Eats' merchant pricing page (Lite 20%, Plus 25%, Premium 30%, self-delivery 15%, pickup 7% with validated in-store pricing), and Grubhub's pricing page (Basic 5%, Plus 15%, All-access 20% marketing commission, with Grubhub Delivery starting at 10%).

Verdict on the tiers: Grubhub Basic at 5% marketing commission is the cheapest marketplace listing on paper, but Grubhub's own footnote says the plans exclude payment processing fees and its delivery fee starts at 10% on top — so a Grubhub delivery order lands in the same neighborhood as DoorDash Basic. If you are on Plus or Premier and cannot point to specific new-customer volume it bought you, you are paying 10-15 points of margin for placement you have not measured.

One more note the sales rep will not lead with: New York City caps these fees by law. NYC's Department of Consumer and Worker Protection enforces caps of 15% to deliver the order, 5% for services other than delivery and payment processing, and 3% to process electronic payment. If you are outside NYC, no cap protects you.

What does the same order earn you through your own site?

A $40 direct pickup order leaves roughly $26 of contribution margin against roughly $17 for the same order on DoorDash Plus — a difference of about $9 per order. Here is the full line-by-line, using a 30% food cost and packaging at $1.25 for delivery and $0.75 for pickup. Swap in your own numbers; the shape does not change.

The $1.29 is not invented. It is Uber Eats' published Webshop rate: a 2.5% order processing fee plus $0.29 per order. On $40 that is exactly $1.29.

Now the part almost every 'ditch the apps' post skips. If you take a direct delivery order and absorb the courier cost instead of billing it to the customer, the math collapses. Uber Direct — same-day delivery fulfilled from your own site — starts at $7.99 per delivery per Uber's pricing page. Subtract that from $25.46 and you are at $17.47, which is $0.72 better than the marketplace order you were trying to escape.

Direct ordering only prints money on two shapes of order: pickup, and delivery where the customer pays the delivery fee (which is exactly what they already do on DoorDash). Chase those two. Everything else is a rounding error.

At what repeat-order rate does direct ordering pay for itself?

The software break-even is effectively zero orders, because the direct ordering surface itself is commission-free — the real break-even is on the marketing that moves customers to it. DoorDash's Online Ordering product is commission-free and can be switched on from an existing Merchant Portal account in under an hour. Grubhub Direct advertises zero setup fees, zero platform fees, and zero monthly hosting fees. Uber's Webshop is 2.5% + $0.29.

So stop asking when the tool pays for itself. Ask this instead: at roughly $9 of extra contribution per shifted order, how many orders do I need to move to cover the demand generation that shifts them?

• 100 orders/month moved from marketplace to direct at ~$9 = about $900/month in recovered contribution
• 300 orders/month moved = about $2,700/month — real money at a 2.8% net margin
• The orders worth moving are your repeats: people who already know your food and do not need the app to discover you
• The orders NOT worth chasing are first-time discovery orders — you are paying the app for the introduction, which is what it is actually good at

Run that against your own POS data. Pull the count of DoorDash orders from customers who ordered from you more than once in the last 90 days. That number, times your own per-order gap, is the size of the prize. If it is under a few hundred dollars a month, keep the listing and go fix something else.

Should you cancel the marketplace listing or keep it for discovery?

Keep it. Cancelling a marketplace listing to save 25% on orders you would not otherwise receive is not a saving — it is a revenue cut. The marketplace is a paid discovery channel that happens to bill as a commission. DoorDash says its plans have no minimum order or sales amount and can be changed or cancelled at any time, so the sane move is a plan downgrade, not a walkout.

The reframe that actually makes you money: treat the app's commission as customer acquisition cost, and judge it the way you would judge any other acquisition channel — by whether it brings you people you did not already have. Then stop paying that acquisition cost a second, third, and tenth time on the same person.

That is the whole strategy. Rent discovery. Own the repeat. If you also want the discovery to come from Google instead of the app — where the click costs you nothing per order — that is what local SEO for restaurants is for, and the restaurant local SEO playbook walks the map-pack side of it.

How do you move a marketplace customer to direct ordering?

You capture them at the moment of highest intent — inside the bag, on the receipt, and on every surface they search before they open the app. There is no clever growth hack here. There is a checklist, and most restaurants do none of it.

• Put your direct ordering link on your Google Business Profile as the order-ahead action link — that is the surface people hit when they search your name
• Insert a printed card in every marketplace delivery bag: a specific offer for ordering direct next time, with a short URL and a QR code
• Make 'Order Online' the single loudest button on your homepage, above the fold, on mobile first — not a link buried in the nav
• Collect the email or phone at direct checkout and use it (this is the only channel where you are allowed to)
• Give direct orders a reason to exist: a free side, a loyalty punch, a lower price — anything the app cannot match, because the app takes 25%
• Never send someone who searched your restaurant by name to a third-party app. That is a paid re-acquisition of a customer you already had

The bag insert is the highest-leverage item on that list and it costs about two cents a copy. Every marketplace delivery is a physical touchpoint with a customer whose contact details you will never be given. It is the only moment you get.

Then the ordering page has to actually convert. Speed matters more than restaurants think: Portent's 2022 analysis of 20 sites and 5.6 million sessions found that sites loading in 1 second converted at roughly 2.5x the rate of sites loading in 5 seconds for e-commerce. If your menu page takes four seconds to render on a phone in a parking lot, you built a very cheap way to lose orders. Our CRO audit methodology covers how to find those leaks.

What does a direct online ordering setup cost to stand up and run?

The ordering software is close to $0 up front — the cost is the site around it and the demand you have to generate yourself. DoorDash's Online Ordering can be activated from an existing Merchant Portal account (their FAQ says setup can take under an hour, or two business days with their onboarding team). Grubhub Direct lists zero setup, platform, and hosting fees. Uber's Webshop charges 2.5% + $0.29 per order.

So the real budget lines are elsewhere:

That last line is the one operators underprice. The 25% commission was never just a payment fee — it was an advertising fee. Take the ad away and you have to replace it. That replacement is a fast, conversion-focused site plus the local search visibility to feed it, which is exactly what our website design service and restaurant SEO work is built to do. Email is the cheapest lever once you own the list — see our take on email tools for service businesses.

Who owns the customer data in each model, and why does it matter?

On the marketplace you get zip codes and frequency buckets. On your own site you get the customer. DoorDash's own Marketplace FAQ answers the question 'What information can I see about my Marketplace customers?' with this: reviews and ratings, zip codes and heatmaps of where customers order from, and which customers are new, occasional, or frequent. That is it. No name you can contact. No email. No phone.

Read that list again and notice what it implies. The platform knows your best customer by name. You know their zip code. Every marketing decision you would want to make — win back the lapsed regular, text the Friday-night crowd about a special, invite the top 50 to a tasting — is unavailable to you and available to them.

A direct order gives you the email, the phone, and the order history, which is the entire raw material of repeat business. That is the asset. The commission is just the rent you pay for not having built it. Getting the Google Business Profile pointed at your own ordering page is the cheapest first brick.

Foundgrove builds the direct-ordering side: a fast site that converts, a Google Business Profile that sends the order-ahead click to you instead of the app, and the local search visibility that replaces the discovery you were renting. Month-to-month, no lock-in, no ranking guarantees. If you want us to look at where your orders are actually leaking, start with restaurant SEO or just Get my free audit.