Own ordering site vs Deliveroo: the real margin maths

What do marketplaces actually cost you?

Food marketplaces are not free to list on and they are not free per order. The headline cost is commission: a percentage of the order value taken before the money reaches you. The current commission band depends on the platform, your plan and whether they deliver or you do, so check your own statements; for this article we will use 30%, the middle of the 25% to 35% of order value that the large platforms are usually reported to take (RTÉ Brainstorm, November 2025), as the rate, and call it C.

Commission is not the whole picture. There are three other costs that rarely appear in the sales pitch.

• Payment processing. Card fees are charged on the full order value, whichever route it comes through. The difference is that on a marketplace they are often bundled into the commission, so you cannot see them.
• Price pressure. Many operators raise menu prices on the apps to protect their margin. Customers notice, and it trains them to see your food as expensive.
• Your customers are not yours. The platform owns the customer record, the email address, the order history and the relationship. You cannot message a regular who orders from you every Friday, because to the app they are its customer, not yours.

What does that add up to in a month?

Put a number on it. Say a typical order is €30. At a commission of C, the platform keeps €30 × C on every order. If C were 25%, that is €7.50 per order; if it were 15%, €4.50. Multiply by the number of app orders you take in a month and you have the figure you are paying for the privilege of being found. For a takeaway doing 300 app orders a month at €30, every five percentage points of commission is €450 a month.

What does a direct-ordering site cost instead?

A direct-ordering site is a website your customers order from directly, built on your own domain, with your menu, your prices and your customer list. Ours is a progressive web app, which means it opens from a phone’s home screen like an app but needs no app store and no download. Our direct-ordering website is a fixed fee from €3,500, built in four to eight weeks, plus a care plan from €60 a month for hosting, updates and backups.

The per-order cost is the card fee, which you pay to a payment provider such as Stripe or SumUp at their published rate; call it F, typically a small percentage plus a few cent. There is no commission. Delivery is either your own driver or collection; many of the operators we work with run their direct site as collection-first, which also removes the delivery headache.

So the comparison is: a percentage of every order for ever, versus a fixed fee once plus a small monthly cost, with card fees on both sides cancelling out.

How many direct orders a month make it pay?

This is the only sum that matters, and it is simple enough to do on a napkin. The saving on each order moved from the app to your own site is the commission you no longer pay:

Saving per order = average order value × C

The site costs its fixed fee once, and its care plan every month. Spread the fixed fee over the period you want it to pay back in. If you want the site to pay for itself within a year:

Monthly cost = (fixed fee ÷ 12) + care plan

Then the number of orders a month that need to come through your own site rather than the app is:

Break-even orders per month = monthly cost ÷ saving per order

Worked through with our illustrative inputs: a €3,500 site and a €60 care plan is €352 a month over a one-year payback. If the average order is €30 and commission C is 20%, the saving per order is €6, and the break-even is 352 ÷ 6, which is about 59 orders a month. That is two orders a day. Every order above that is margin back in your pocket, and in year two the fixed fee is gone and the break-even drops to ten orders a month.

Redo it with your own order value and your real commission rate. If your commission is higher, the break-even is lower. If your average order is bigger, lower again.

How do you move regulars from the app to your own site?

A direct-ordering site only pays if people use it, and nobody will find it on their own. The good news is you do not need to win new customers to make it work; you need to move the ones you already have. Regulars are the easiest people to move because they already trust the food.

• Put the offer in the bag. A card in every app order: ‘Order direct next time at yourname.ie and get free garlic bread.’ A giveaway worth €2 to win back €6 of commission is a trade you make all day.
• Make direct cheaper, not just the same. If your app prices are inflated to cover commission, show the honest price on your own site and say so.
• Collect contact details on your own site and use them. A short message on a quiet Tuesday to people who have ordered before is the highest-return marketing a takeaway can do.
• Fix your Google Business Profile. When someone searches your name, the ‘Order online’ button should go to your site, not the marketplace.
• Tell people in person. Staff at the counter, a sign at the till, a line on the receipt. It is slow and it works.
• Reward loyalty on your site. A simple points or stamp system that only works direct gives regulars a reason to switch and stay switched.

Should you leave the marketplaces altogether?

Usually not. The apps are expensive for the orders you would have got anyway, but they are good at one thing: putting your food in front of people who have never heard of you. That is discovery, and it has a value. The mistake is paying discovery prices on every order from every customer for ever.

The pattern that works is to stay listed, treat the app as advertising, and move each new customer to direct ordering as quickly as you can. Over a year the mix shifts. Your app orders become mostly first-timers, your direct orders become mostly regulars, and your blended commission rate falls every month.

There is one exception. If you are in a very competitive area and the app is also doing your delivery, the maths is different and you should run the formula above with the delivery cost on the direct side. For collection-led businesses, the case for direct ordering is almost always clear.

What did Pesto Kitchen do?

Pesto Kitchen is a Neapolitan pizza business at Elmgrove Farm in Gormanston, Co. Meath. We built their site with direct online ordering as a progressive web app, so customers order from a browser, save it to their home screen and come back without an app store in between. The menu, the prices and the customer relationship belong to the business.

The results from the first period are being agreed with the client before we publish them. The full write-up, including what we would do differently, is in the case study.

If you want to run the formula with your own numbers and see where you land, download our free guide Taking orders online without marketplace fees, or book a free 30-minute call and we will do the maths with you.