Delivery apps vs your own list: who owns your customer?

The order comes in on the tablet, the kitchen fires it, a driver picks it up, and the money lands minus the commission. Somewhere in your city, a person just ate your food and liked it. You don't know their name, their number, or whether that was their fifth order or their fiftieth. That's not a technology gap. That's the business model.

This isn't a sermon against delivery apps. They solve real problems and they've earned a place in most restaurants' week. But there's a structural difference between a customer on a marketplace and a customer on your own list, and that difference decides who profits from the relationship over time. It's worth seeing clearly before you decide where to push.

Three things the marketplace keeps

Start with what actually happens when someone orders your food through a third-party app. Three things stay on the platform's side of the table, permanently.

• The commission. Marketplaces take a cut of every order, and depending on the plan it can reach 30 % of the ticket. Not just on the first order, while they're introducing you. Every order, including that customer's fiftieth.
• The customer data. You get an order to fulfil, not a person to know. Name, number, ordering history, favourite dish: that file exists, but it belongs to the platform. You can't thank the customer, tell them about a special, or notice when they stop ordering.
• The rules. Commission rates, ranking in search, service fees, refund policies. The platform can change any of it, and your only lever is to accept or leave. And when your regulars live inside someone else's app, leaving costs you your regulars.

None of this is hidden, exactly. It's the deal: the platform brings demand and keeps the relationship. The real question is whether you want that to be the deal for your best customers, forever.

The rent, in real dollars

Run the numbers on one loyal customer. Say an order runs $40 and the commission is 25 % — an assumption for the sake of the math; your actual rate depends on your plan. That's $10 per order to the platform. If that customer orders twice a month, that's $240 a year in commission, for one person, every year, with no loyalty discount for either of you. Ten customers like that and you're paying $2,400 a year to keep serving people who already chose you.

Now compare a direct customer on your own list. The margin stays on the ticket. Reaching them costs you a text message. And when they drift, you can see it and do something about it. The delivery customer who drifts just disappears, and you never find out it happened.

When the apps genuinely make sense

Here's the honest case for the marketplaces, because there is one.

• Discovery. A new customer scrolling at 7 p.m. finds you because you're on the platform. That first order is one you'd likely never have gotten otherwise. Treat the commission on it as a marketing cost and it's often a fair price.
• No fleet of your own. Hiring drivers, insuring them, dispatching them: most independents can't and shouldn't. The apps solved a logistics problem you don't want to own.
• Slow nights. Incremental orders on a Tuesday, when the kitchen is staffed anyway, can carry a commission and still add margin.

The problem was never the first order. It's the tenth, the twentieth, the customer who found you two years ago and still pays platform rent on every visit to your menu. Discovery is worth a commission. Loyalty is not something you should have to keep buying back.

Turning app customers into direct regulars

You can't message the platform's customers, but plenty of them physically show up: pickup orders, the couple who ordered delivery last week and walked in tonight, the office nearby that found you on the app. Every one of those moments is a chance to move the relationship to your side of the counter.

The mechanics have to be fast, because the counter at pickup time is no place for paperwork. With Perkaria, joining takes a few seconds: your staff enters a first name and a mobile number, and the customer's card arrives by text. No app to download, no password, no form on a clipboard. How it works is deliberately boring — a tablet beside your register, two taps, next customer.

• Give the counter a one-line script: "Want your visits to count? Takes a few seconds — what's your cell number?"
• Put the offer where pickup customers stand. A small sign at the pickup shelf beats a flyer nobody reads.
• Make the first reward feel reachable. A stamp card with a reward at ten visits gives an app customer a concrete reason to order direct next time.
• Train it once, then trust it. The counter app is two taps behind a staff PIN, so the new hire runs it on day one.

You won't convert everyone, and you don't need to. Every customer who moves from the app to your list stops paying rent on their own loyalty, and starts compounding for you instead.

A list you own, built the legal way

A first-party list is only an asset if it's consented. In Canada, CASL requires consent before a commercial message, clear identification of the sender, and a working unsubscribe in every message. That's not red tape to route around; it's what makes the list worth having. A hundred people who said yes will beat a thousand who didn't, because the hundred actually read what you send.

This is also where discipline pays. Campaigns in Perkaria run inside limits the platform enforces for you: nothing promotional leaves between 9 p.m. and 9 a.m., no member gets more than two marketing texts a week, and STOP or ARRET removes someone the moment it arrives. Those aren't just compliance features. They're what keeps a text from your restaurant feeling like a note from a regular's favourite spot instead of noise from a brand.

What ownership looks like in practice

Owning the list isn't just about sending. It's about seeing. Once your regulars are members, the questions you can't answer about app customers become dashboard rows: visits per member per month, your top customers by name, the regulars who quietly stopped coming in. Insights turns those into a Monday-morning digest, and the lapsed list feeds a win-back message that goes only to people who actually drifted.

That's the whole loop. The apps hand you strangers; the counter turns some of them into members; the list keeps them; the numbers tell you when one is slipping. Each step is small. The compounding isn't.