Run the numbers on a $12.95 burrito. A delivery platform takes something in the range of 15 to 30 percent depending on your tier, so the same item listed at your menu price nets you roughly $9 to $11. Price it at $15.95 in the app and you land near where you started. Almost every restaurant on those platforms has done some version of that arithmetic.
Which means most restaurants operate two menus at two price points, and only one of them is under your control on the phone.
The moment it shows up on a call
A customer searches your restaurant, lands on a platform listing because platform listings rank well, reads the marked-up menu, and calls you directly to skip the delivery fee. Then they say a number back to you that is four dollars higher than your menu.
Whoever answers has about two seconds to decide what to do, and there are three common outcomes. They honor the app price, which quietly overcharges a customer who called you specifically to give you more margin. They get flustered and explain commission economics, which converts a thirty-second order into a two-minute grievance session about DoorDash. Or they say the useful sentence: "That's the app's price. Ours is $12.95."
The third answer is correct and it takes four words more than the first. It should not require judgment from a nineteen-year-old holding a phone against their shoulder, but with no script it always does.
Why staff instinctively honor the higher number
Because correcting a customer's price feels like arguing, and every server has been trained out of arguing about price. The instinct is protective, not lazy. It is also backwards here, since the correction is in the customer's favor.
Name the situation explicitly in training. Callers will quote app prices, that is expected, and the direct price always wins. Say it as a rule so nobody has to improvise it during a rush.
Can you legally price differently
Mostly yes, with two things to check.
The first is your merchant agreement. Some platform contracts historically contained price parity provisions requiring in-app prices to match your own. Those drew antitrust attention and legislative pushback, and several were narrowed or dropped. Whether one is still in yours depends on when you signed, which platform, and what has been amended since. Read the current document. If you cannot find it, ask your account manager in writing and keep the reply.
The second is your own disclosure obligation on the direct channel. Charging more on a platform is a platform pricing decision. Charging a caller more than the price you quoted them is a different thing entirely, and the analysis for that sits in junk fee disclosure rules for restaurants. The parity question and the disclosure question get blurred together constantly, and they are not related.
A handful of cities have legislated on delivery commissions, and a few of those measures touch pricing in ways that vary by jurisdiction. If you operate in a city that capped commissions, ask your counsel whether anything in that ordinance reaches your menu pricing. Elsewhere, this is a contract question rather than a legal one.
What a two-price structure costs you that is not obvious
The commission math is the visible cost. Four less obvious ones do more damage over a year.
- Search results show a customer your inflated price before they ever see your site, so your perceived price point is set by a listing you did not design.
- Every menu change now has to happen in two places, and the platform copy is the one that drifts, because it is nobody's job in particular.
- Pickup arguments happen when a customer ordered through the app for pickup, paid the marked-up price, and sees the printed menu on your counter.
- The price difference erodes trust with your best repeat customers, who order both ways and notice, even when they understand exactly why the gap exists.
None of these is fatal. Together they are an argument for keeping the direct price loudly and consistently visible on every channel you control, so the platform number reads as the exception rather than the baseline.
The drift problem deserves its own attention, because it compounds quietly. You raise a price in the POS on a Tuesday. The printed menus get reprinted at the next order, six weeks out. The website updates whenever your web person gets to it. The platform listing updates when somebody remembers there is a platform listing. Six weeks later you have four prices for one item and no single source of truth, and every customer-facing channel is confidently quoting a different one.
Restaurants that handle this well tend to do one boring thing: they treat the POS as the price of record and give one person the job of pushing every change outward on the same day. Not a system, a person and a checklist. The operators who try to solve it with more software usually end up with a fifth price.
The phone is the one channel where you can fix this cheaply
The phone quotes your price by definition. There is no platform between you and the caller and no commission on the ticket. That is the whole argument for pushing volume back to it, examined more fully in voice AI vs delivery app commissions.
A voice agent reads your own menu, so it quotes your own prices on every call without anyone deciding to. It never honors an app price out of politeness and never gets drawn into an argument about platform economics. Ask it about the price difference and a well-configured agent gives the short version and moves on.
The failure mode moves rather than disappearing. A human quoting from a stale printed menu catches themselves eventually, because they ring the item up and see a different number. An agent quoting from a stale menu file does it identically on every call, all day, with complete confidence, and nobody catches it until the tickets stop matching the payments. Menu sync stops being a nice-to-have the moment a machine is doing the quoting, which is why menu sync and real-time 86ing deserve more attention than their dull names suggest.
If you are also pulling platform orders into the same ticket flow, the price-source question gets sharper still. Two order streams with two menus landing on one kitchen rail is how a store ends up selling a $12.95 burrito for $15.95 to the wrong customer. The mechanics of keeping those straight are in consolidating third-party delivery phone orders.
Here is the test worth running. Open your platform listing on a phone, put your own printed menu next to it, and compare every item. Then call your restaurant and ask the price of three items you deliberately changed in the last six months. If the platform list and the printed menu and the spoken answer are three different numbers, the price gap is no longer a pricing strategy. It is a data problem wearing a pricing strategy's clothes, and the customer standing at your counter with a screenshot is the one who found it first.