Move Your Regulars Off Delivery Apps, In Order

You already know what the apps cost you. If you have not run the math on your own numbers, start with why your delivery app orders cost more than you think and come back. This post is the part after that: the actual moves, in the order they work, to get the people who order from you every week to order directly instead.

A few things to set expectations. You are not going to move everyone, and you should not try. Some customers are app customers — they found you by scrolling, they have credits, they will never care who you are. Leave them. The people worth moving are the regulars: the ones who order the same thing every Thursday, the ones whose address your driver knows. Those orders are the profitable ones, and they are the ones most likely to move, because those customers already like you specifically.

And this is slow. It happens one bag at a time. Give it a full quarter before you decide whether it is working.

First: make sure direct ordering is actually better

Do not start marketing your own ordering until it is genuinely the easier option. If a customer tries your link once and it is clunky, you have spent the only attention they were going to give you.

Open your own ordering page on your phone, standing up, on data, not wifi. Order the thing you sell most. Count the taps and the seconds. Then do the same order on the app. If yours is slower, fix that before you print anything.

The specific things that usually make direct slower:

• A forced account signup. Guest checkout or nothing. You can collect the phone number at checkout anyway — that is the account.
• A menu that loads as one enormous page. Categories at the top, jump links, images that are actually sized for phones. Pages that take a beat to load lose people at a rate that surprises owners; we wrote about that in website speed.
• Address typing. Autocomplete on the address field is not a luxury. Typing a Toronto postal code on a phone keyboard is where orders die.
• No saved reorder. The single biggest advantage you can build over an app is a one-tap "order that again" for people who buy the same thing. Regulars are, by definition, repeaters.
• No clear pickup time or delivery window. People will not commit to an unknown.

If you do not have ordering on your own site yet, or you are running on a PDF menu and a phone number, that is the actual first step — see our restaurant work for what a direct ordering setup involves, and why a menu PDF is not an ordering system.

Second: match the price, or beat it slightly

This is the one owners resist and it is non-negotiable. If your app menu is marked up 20 percent to cover commission and your direct menu is the real price, good — that is already an advantage and you should say so. But if your direct price is higher than the app price for any reason, or your direct delivery fee is worse, nothing else in this post will work. Customers check. They are on a phone with two tabs open.

The clean version: same food price everywhere, and direct gets something the app cannot match. Free pickup. No service fee. A lower delivery minimum. Pick one, keep it permanent, and make it a sentence a customer can repeat to a friend.

What you should not do is get into a discount war with yourself. A standing 20 percent off direct orders trains people to only order when they feel like they are getting away with something, and it eats the margin you just recovered from the app. The advantage should be structural — no fees — not a coupon.

Third: the bag insert

Every app order that leaves your kitchen is a delivery you are paying for. Put something in the bag.

Not a flyer. A flyer reads as advertising and goes in the recycling with the receipt. What works is a small card, stapled to the bag or taped to the lid, that says three things:

1. Thank you, by name if the app gives you one.
2. Order direct next time at your short URL — and a QR code, big enough to scan from a countertop.
3. The one concrete reason: "same prices, no service fees" or "pickup is always free direct."

Keep it to those three. No mission statement, no history of the restaurant. The card is being read by someone who is hungry and about to eat, and you have roughly four seconds.

A practical note: make the URL short and typeable, because plenty of people will not scan a code. yourplace.ca/order is fine. yourplace.ca/online-ordering-system/menu is not.

Fourth: the one sentence at the counter

Pickup and dine-in customers are the easiest to move and most restaurants say nothing to them at all.

One sentence, said by whoever hands over the bag, every time: "If you order from us again, order right on our site — it's the same price and we get the whole thing."

That last half matters more than any discount. People in this city know the apps take a cut. Saying it plainly turns the ask into a small favour instead of a sales pitch, and regulars will do you a small favour. It is the same mechanic as asking for the Google review while they are still standing there — the request works because it is specific, timed, and easy to say yes to.

Train it like you train anything else on the floor. Write the sentence on a card by the pickup station. Say it in the pre-shift for two weeks until it is automatic.

Fifth: make the first direct order feel noticed

When someone orders directly for the first time, do something small and human. A handwritten "thanks for ordering direct" on the receipt. An extra piece of whatever you make that travels well. A note in the bag.

The point is not the item. The point is that the customer now has a reason to believe direct is a different experience than the app — a place where someone knows they came. That is the only durable advantage you have over a platform, and it costs you almost nothing.

Then do one more thing: when the order comes in with a phone number, save it. Tag the customer. If you have a simple text list, ask at checkout with a checkbox, not a pre-ticked one. A short list of people who actually want to hear from you beats a big list that marks you as spam.

How to tell if it is working

Here is where most owners measure the wrong thing. Raw direct order count goes up whenever you run a promo and back down after, and it tells you nothing about whether you have actually moved anyone.

Measure repeat rate on direct orders. Specifically:

Of the customers who placed a direct order in a given month, what percentage placed another direct order within the next 60 days?

Pull it by phone number or email, whichever your system stores. Do it monthly. It will be a small number at the start and that is fine — what you want is the direction. A rising repeat rate means the switch is sticking: people tried direct, liked it, and came back without being asked. A flat repeat rate with rising order counts means you are buying one-time orders with discounts, and the moment you stop the discount it stops.

Two supporting numbers worth a glance each month:

• Direct share of total orders. App orders plus direct orders, what fraction is direct. Watch the trend over a quarter, not week to week.
• Average direct ticket versus app ticket. Direct usually runs higher once regulars move, because the people who move are the people who order more.

If after three months the repeat rate has not budged, the problem is almost always in step one. Sit down with your own ordering flow and a stopwatch again. Something in it is asking your customer for more effort than the app does, and no card in a bag will beat that.

The order matters

Fix the flow. Match the price. Put a card in every bag. Say the sentence at the counter. Notice the first direct order. Then measure repeat rate, monthly, and be patient with it.

Doing these out of order is why this usually fails. Owners start with the bag insert because it is the easiest thing to print, send people to an ordering page that takes ninety seconds and three screens, and conclude their customers just prefer the app. They do not. They prefer whatever is faster at the moment they are hungry. Make that be you.