"Commission-free" might be the most abused phrase in restaurant tech. Every vendor selling ordering software prints it, and every owner who's been burned once reads it as code for "the costs are hidden somewhere else." That's a reasonable thing to assume. So skip the pitch. Here's the full invoice: every dollar you pay to run your own ordering channel, set next to the marketplace bill it replaces.
Why the two bills behave differently
A marketplace takes a percentage of everything, forever. The published tiers in 2026 run 15%, 25%, and 30% (DoorDash's Basic, Plus, and Premier as of mid-2026, with Uber Eats spanning the same 15–30% after its March 2026 increase). Add promo participation, processing, and refund chargebacks, and operators regularly land at an effective 30–40% of gross. The bill scales with your success. Double your sales and you double what you hand over.
Your own channel works the other way around. Most of the cost sits fixed (software, a little marketing), and the variable part stays small: payment processing and whatever delivery costs you. Double your sales here and the cost per order actually goes down. That one difference decides everything below.
The four lines you actually pay
First, the software. A white-label ordering platform (an ordering website, a branded app, an admin panel) runs a few hundred dollars a month for a single location. Call it $400 for the math below. DIY plugins cost less and custom development costs 100× more; most restaurants land in the middle, and we break the whole range down in what an online ordering system costs in 2026.
Then payment processing. Nobody escapes this one; you pay it on marketplace orders too, directly or baked in. Standard card rates run 2.9% + $0.30. On $20,000 of monthly orders at a $30 average check, about 667 orders, that comes to roughly $780.
Delivery is next. If customers pick up, it's zero. If you run your own couriers, operators report $3–6 per drop in most US suburbs once routing and auto-assignment are set up, and a delivery fee passed to the customer usually covers most of it. This is the line that varies most, so run it on your own zone map rather than someone else's average.
Last, marketing the channel — the honest line most "commission-free" pitches skip. A marketplace brings traffic; your own site doesn't sell itself. Budget for bag inserts, a Google Business link, and retention campaigns: $300–500 a month covers a serious effort at one location.
Same $20,000 in orders. The marketplace takes $6,000. Your own channel runs about $1,680 plus delivery. Even if your couriers eat another $1,500, you come out $2,800 ahead every month.
Bring last month's marketplace statement. We'll do the math together in 30 minutes.
The break-even is lower than most owners guess
Because your costs are mostly fixed, there's a volume below which the marketplace really is cheaper. To find it, divide your fixed monthly cost by the commission rate. At $900 fixed (software plus marketing) against a 30% cut, break-even lands at $3,000 a month in direct orders, roughly 100 orders, or three to four a day. Below that, don't bother. Clear it and every extra order widens the gap in your favor.
Most restaurants doing meaningful marketplace volume clear that bar with the regulars they already have. Those customers exist; they're just ordering through a middleman that charges you 30% for the introduction every single time, including the fiftieth order. Moving even a fraction of them over is the playbook in our guide to migrating delivery-app customers.
What the fixed cost buys beyond the savings
There's more to owning the channel than the fee gap. The conversion math changes too: across 3M+ orders processed on Dots, branded apps convert up to 35% of visitors and ordering sites up to 15%, versus roughly 3% for a listing buried in a crowded marketplace. Traffic turns into more orders when it lands on a page that shows one restaurant, yours.
You also keep the customer record. Name, phone, order history: the raw material for cashback, push campaigns, and win-back flows that run without ad spend. On a marketplace all of that stays with the platform. And going live doesn't take a year of development; a white-label setup is up in about two weeks, menu and payments included.
The two objections that come up on every call
One is "my customers won't pay a delivery fee." They already do. Marketplaces charge them service fees and delivery fees on top of your marked-up menu. A $2.99 flat fee on your own site, with honest menu prices 15–20% below your marketplace listing, still comes out cheaper for the customer on almost every order. Put the comparison on your site; people can do the arithmetic.
The other is "I don't have time to run another system." Fair, if it really is another system. The test is whether orders, couriers, menu, and marketing live in one admin panel or four browser tabs. One panel costs you minutes a day; four tabs cost you a part-time job. Ask the vendor to walk through the daily workflow end to end, not the feature list.
Do this before you sign anything
Pull last month's total marketplace deductions. Above $1,500, commission-free ordering pays for itself even on pessimistic assumptions, and the only open question is execution. Under $500, keep the marketplace as your storefront for now and revisit once volume grows. A vendor who won't sit down and work through this break-even with you is telling you something.
Branded app, ordering site, logistics, and marketing tools — live in about two weeks.