Delivery apps are convenient, and they are quietly eating restaurants alive. Commissions of twenty to thirty percent on every order mean that a busy month on the apps can still be a thin one on the books. The apps own the customer relationship, set the rules, and take their cut whether you have a good month or a bad one. For a lot of restaurants, that is the single biggest leak in the business.
The fix is not to abandon the apps overnight, it is to give customers a reason to order direct. A restaurant website with proper online ordering lets a regular skip the app, order straight from you, and save you the commission on every single order they place. The customers who already love you are exactly the ones who will happily order direct if you make it easy.
What restaurant delivery apps really cost you
The commission is the part everyone sees. The quieter costs are the ones that compound. When someone orders through an app, the app gets their name, their number and their order history. You get a ticket to cook. If that customer never orders again, you have no direct way to invite them back, except by paying the app to show them your listing again.
The app also decides how you appear. Your position in the list, the photos beside your name and the promotions running next to you all belong to someone else. A competitor can sit one scroll away from your menu, on a screen you do not control. On your own site, the only menu on the page is yours.
None of this makes the apps useless. They are genuinely good at one thing: putting you in front of people who have never heard of you. The mistake is letting them handle your regulars too. A first order through an app is marketing. The tenth order from the same person through the same app is a cost you chose to keep paying.
To see your own number, add up a month of app statements: commissions, marketing fees, payment fees, refunds charged back to you. Divide that by the month's app sales. That percentage, not the headline rate, is what each app order really costs.
What a direct-ordering site actually needs
It needs to be fast and effortless on a phone, because that is where the decision to eat gets made. It needs a menu that loads instantly and looks good enough to make someone hungry, not a blurry PDF. And it needs ordering and payment that take a couple of taps, because every extra step sends people back to the app they already have installed.
Speed deserves its own mention. A hungry person on a phone does not wait for a heavy page; they go back to the app that opens instantly. The three-second rule for page loads applies to menus as much as any other page, and a slow ordering site quietly sends customers back to the commission you were trying to avoid.
Before I call a restaurant site ready, I check it against a short list:
- The menu is real text on the page, not an image or a PDF, so it loads fast, reads well on a small screen and can be found by search engines.
- Prices and availability are current. Nothing kills trust faster than ordering a dish that turns out to be off the menu.
- Checkout works without an account. Let people order as a guest, and offer an account afterwards to those who want one.
- Payment options match your customers. Cards, wallets and, where it is normal locally, cash on delivery.
- Hours, delivery area and phone number are easy to find on every page, without hunting.
On the technical side, describe your menu, hours and address with structured data based on the schema.org Restaurant type. Search engines then read those details from the page instead of guessing them.
Reservations matter too. The same site that takes orders can take bookings, filling tables without the phone tag and capturing customer details you actually own. Over time those details let you bring people back directly, instead of renting access to your own customers from a third party.
How to move regulars to direct ordering
Building the site is half the work. The other half is telling people it exists, and the best audience is the customers you already have.
- Put your website address on everything that leaves the kitchen. Packaging, receipts and a small card in the bag all reach people who already like your food. Read your agreement with each app first, because terms differ.
- Give a reason to switch. A small standing benefit for ordering direct, such as a free side or a loyalty stamp, can cost less than the commission it replaces. Work it out against your own menu and margins before you pick one.
- Point your Google listing at your own ordering page, so people who search for you by name land on your site and not on an app.
- Use the details you collect, with permission, to tell regulars about new dishes or quiet-night offers.
Mistakes that send customers back to the apps
- An "Order online" button on your own website that links straight to a delivery app.
- A direct menu that costs more, or takes longer to use, than the app version.
- Opening hours that are out of date, so people order and then find you closed.
- A checkout that forces account creation before showing the total.
Website first, app later (if ever)
Some owners ask whether they should build their own mobile app instead. For most single restaurants and small groups, a fast website that works well on phones does the job. A dedicated app only starts to make sense when enough customers order often enough to keep it on their home screen. I explain how to tell the difference in do you need a website, a web app or a mobile app.
The same idea, removing every step between a ready customer and a sale, applies outside food. Clinics lose patients to friction in much the same way, and why your clinic website isn’t booking appointments walks through the fixes there.
The maths is simple: every direct order is the same revenue without the commission. A website that makes ordering direct as easy as the apps pays for itself fast. I cover how I build these on my restaurant website design page, but the goal is always the same, to own your customers instead of renting them.
