Getting listed
How to get your restaurant onto delivery apps
Getting a restaurant listed on Uber Eats, Glovo or Bolt Food takes three to six weeks, and most of that is the platform's own paperwork moving at the platform's own speed. Your part is smaller than people expect: one page of company details, a dish list with prices, and a folder of photos. This page walks through the order it actually happens in, how long each step really takes, and the handful of things that quietly add a fortnight.
The order it happens in
Nobody publishes this sequence, which is why so many restaurants stall halfway. Here it is, with honest durations. The steps overlap — that is the whole trick to doing it in three weeks instead of eight.
-
01
Pick the apps and start both at once
1 dayTwo platforms, opened in parallel. Not one first and the other later — see the next section for why.
-
02
Get in front of a platform manager
3–7 daysYou can fill in the public sign-up form and wait your turn in the queue, or you can be introduced by someone who already talks to that manager weekly. The second route is faster and it is the only one where you get to ask questions before signing.
-
03
Sign the contract
1–2 weeksThis is the long pole, and it is not in your hands. A bilateral agreement, their template, their legal team's pace. Chasing it politely helps. Nothing else does.
-
04
Build the store
3–7 daysMenu, categories, item names, descriptions, prices, modifiers, combos, photos, opening hours, preparation time. This can and should happen while the contract is still moving, so the store is ready the day access is granted.
-
05
Train the kitchen, place a test order, go live
2–5 daysSomeone on your team learns four buttons. Then a real order goes out to a real address so you can see what the food looks like after twenty-five minutes in a bag. Then the platform reviews the store and switches it on.
-
06
Advertising and the daily grind
ongoingCampaigns, photo and name tests, ratings, replies to reviews. From here it stops being a project and becomes a routine.
Three weeks is the good case: paperwork ready on day one, someone building the store in parallel, a manager who answers. Six weeks is the normal case.
Why the apps say “a few days” and it takes six weeks
Read the platforms' own pages and you get a much cheerier picture. Uber Eats says it is possible to become an Uber Eats restaurant partner and start accepting orders in just a few days
. Bolt Food promises you can start selling on Bolt Food in a week
. Then you talk to restaurants who did it and hear six weeks. Everyone is telling the truth. They are timing different things.
The platforms are timing their own part — from a completed application to a store that is switched on. Application, verification, menu upload, live. If your paperwork is in order, your menu exists as a document and there is nothing to negotiate, that really can be days.
You are timing the whole thing — from “we should probably be on delivery” to money arriving. That includes finding your registered address, waiting on a legal team, deciding what a burger costs on an app, renaming two hundred photos, and discovering the soup does not travel. None of that is on the platform's clock, and all of it is on yours.
So plan for six weeks and be pleased when it is three. The gap between the two numbers is almost entirely made of decisions nobody has made yet — which is good news, because decisions are the part you can make faster.
Start on two apps, not one
The instinct is to try one app, see how it goes, and add the second later. It costs you months. The apps do not share an audience: in most European cities one skews to expats and English-language search, another to locals who arrived through a discount, and a third has less traffic but almost no competition in your category. Being on one of them means being invisible to the other's customers, and they are not the same people.
There is a second reason, and it is the one that matters more. Every platform's algorithm judges a new store on its first few weeks: conversion, cancellations, preparation time, rating. That clock starts the day you go live. Running two clocks at once means that in six weeks you have two stores with history instead of one — and history is what the search results are built from.
The practical order: open the two biggest apps in your city together, add the third once the first two are steady. The third one is nearly free to run at that point, because the menu, photos and prices already exist and only need copying across.
What they will ask you for
Here is the whole list. It genuinely fits on one page, and it is the same list on every platform give or take a field. Gather it once, in one document, and the rest of the process stops being about you.
Tick things off as you find them — nothing is saved anywhere, this is just so you can see what is left.
Company details
Operations
Menu and prices
Photos
That is the entire list. Item names, descriptions, category order, keywords, the cover image, advertising, review replies — none of that belongs on it, because none of it is information only you have. It is work, and work can be done by someone else.
One thing on that list is needed earlier than people expect: the menu itself. Uber Eats asks for a copy of your menu as part of the application, not after approval, and Bolt Food reviews the menu before switching the store on. A menu that exists only in someone's head is enough to stall the very first step.
How orders reach your kitchen
You will be asked to pick one of three, and the honest answer is that it depends on your counter, not on which is technically best.
A tablet from the platform. Some platforms will hand you one with their app already on it. Useful when the kitchen has no spare device and nobody wants to own one. It also means one more thing on the counter that only does one job.
Your own tablet or phone. The most common choice. The app installs on Android or iOS, and if you are on two platforms you end up with two apps on one device, which is fine. Keep it plugged in. A dead tablet at 8pm looks exactly like a closed restaurant.
The browser on the till computer. Orders arrive in the web back office. Works well if there is already a screen someone is looking at anyway, badly if that screen is in an office at the back.
Straight into your POS. If you already run a till system, orders can drop into it directly instead of onto a second screen — Uber Eats names Clover, Square and Toast among the systems it plugs into. This is the best option if you have one of them, because it means one screen and one set of tickets. It is also the one that takes longest to set up, so it is usually done after launch rather than before.
Whichever you pick, the sound has to be audible from where the food is made. The single most expensive setup mistake is a device that pings in a room nobody is standing in — every minute it sits unanswered is counted against your preparation time, and preparation time is one of the things the app ranks you on.
Set your prices before you open
This is the one part you cannot delegate and cannot postpone, because the store goes live with whatever prices are in it. The short version: the platform takes a commission on the price the customer pays — usually somewhere between 25 and 30 per cent in Europe, with your own rate written into your contract — while your food costs exactly what it cost before. Copy your dine-in prices across and that commission comes straight out of your margin.
So delivery gets its own price list, normally 25–30% above dine-in. Start at the lower end: a new store has no rating and no position, and being cheap enough to try beats being correctly priced and ignored. Raising a price in three months is a five-minute job.
Two things people forget at this stage and regret later. Packaging goes into the food cost, per dish, not as a menu-wide guess — a box, lid, bag and napkin add 40–60 cents and that is real money by the three-hundredth order. And promotions come out of your price, not the platform's commission, so the dishes you intend to discount need another 10–15% built in from day one, on those items only.
We wrote the arithmetic up properly, with a calculator you can put your own numbers into, in how to price your menu for delivery apps. Do that before you send anyone a price list.
What actually delays a launch
Not the platforms. In our experience four things cause nearly every slipped launch, and all four are avoidable in an afternoon.
Company details arriving one field at a time. The tax number on Monday, the IBAN on Thursday, the registered address the following week because it turned out to differ from the restaurant address. Every missing field is another round trip through someone's inbox. Fill in the whole block above before the first conversation and this stage disappears.
Photos called IMG_4471.jpg. Two hundred files, no way to tell which is the burger. Renaming them is nobody's favourite evening, but it is an evening, and without it the store cannot be built at all.
A menu that has never been in a bag. Something leaks, something goes soggy, something arrives in pieces. You find out on the first real order, from a one-star review, and now you are fixing the menu and the rating at the same time. Send one order to a real address before you go live — it is the cheapest test in the whole process.
Nobody owning the device. "The team knows" is not the same as one named person whose job it is. Shifts change. The person who was trained is off on Saturday. Name someone, and make sure every shift knows the four buttons, not just the one who happened to be there.
What the first month looks like
Manage your own expectations here, because the first weeks look worse than they are and plenty of restaurants panic and start changing things at exactly the wrong moment.
The algorithm is studying you. A new store gets shown to a small audience first while the platform watches how people behave: do they order after opening the menu, do you cancel, how long do you really take. Two to four weeks of run-up is normal. Nothing is broken.
Advertising costs more at the start. With no order history behind you, every click is priced higher. It settles. Budget for the first month being the least efficient one you will have.
A single day tells you nothing. Eight orders yesterday, two today — that is weather, a public holiday, roadworks, or how many couriers happened to be online. Look at a week. Preferably a month. The one number worth watching daily is whether the store is actually open in the app, because platforms close stores automatically after cancellations and nobody tells the kitchen.
Doing it yourself: where the forms actually are
If you would rather handle this without anyone in the middle, that is a perfectly reasonable choice and here is where each one starts. These are the platforms' own pages, not ours.
Uber Eats
Public application form. Asks for your location, a licence and a copy of your menu, then verifies the business before you upload anything.
Glovo
Registration goes through the partner portal. The public page is thin on detail — the requirements arrive in the conversation that follows, not before it.
Bolt Food
States its conditions openly: a registered business meeting local rules, a bank account, and premises to cook in. Says a week from registration to selling, and will send you a device or let you use your own.
What none of those pages will do is tell you what to charge, write item names that get found, or say which of your dishes should not go on the app at all. That part is not in the form.
Questions people ask
How long does it take to get a restaurant on a delivery app?
Three to six weeks from the first conversation to the first order. The platforms quote less — Uber Eats says a few days, Bolt Food says a week — because they are timing their own part, from a completed application to a store being switched on. The restaurant is timing everything around that as well: company paperwork, delivery prices, photos, and the contract, which is the longest single step at one to two weeks and moves at the platform’s pace rather than yours. Building the store takes three to seven days and can run in parallel with the paperwork, which is how the fast version stays fast.
What documents do delivery platforms need from a restaurant?
The registered company name and tax number, an IBAN for payouts, the restaurant address and the registered address if they differ, and a named contact with phone and email. An operating licence and food-safety paperwork are not always requested but are worth having ready. Beyond documents they need your opening hours, a realistic preparation time, and the menu with final prices.
Should I launch on one platform first or several at once?
Several at once, normally two. The apps have different audiences, so one platform reaches roughly half your potential market. More importantly, every platform judges a new store on its first weeks of conversion and reliability — starting both clocks together means that in six weeks you have two stores with trading history rather than one.
How much commission do delivery apps charge restaurants?
In Europe the commission is usually somewhere between 25 and 30 per cent of what the customer pays, and it varies by platform, by country, by service level and by what was negotiated. Your own rate is in your contract, and it is the only figure worth using in any calculation — the published rate is a starting expectation, not what appears on your invoice.
Can I use my dine-in prices on delivery?
You can, and it is the most common expensive mistake. The commission is charged on the price the customer pays while your food cost stays the same, so without a mark-up the commission comes out of your margin. Most restaurants run a separate delivery price list at 25–30% above dine-in, with a further 10–15% on the specific items they plan to discount later.
Do I need professional photos to start?
Not to start, but they are the fastest lever you have. The cover image decides whether anyone opens your store in a list of thirty; the dish photo decides whether they order. Begin with the best pictures you have, named after the dishes, and plan one proper shoot once orders are coming in. Weak photos can be improved; missing ones cannot.
Why did my new store stop getting orders after a good first week?
Check first whether the store is actually open in the app. Platforms close stores automatically after cancellations or a missed order run, and the app shows nothing to the kitchen — staff simply notice it has gone quiet. After that, the usual causes are a rising preparation time, items sitting in the out-of-stock list, or a rating that dipped below the band the algorithm favours.
Where this comes from
- The durations
- Taken from our own onboarding of restaurants in Lisbon on Uber Eats, Glovo and Bolt Food. They are what we plan against and what we have seen repeat, not a published service level from any platform. Your city and your manager will vary.
- The commission figure
- The 25–30% band is a general European range, given so you can do rough arithmetic before you have a contract. We do not publish the rate of any named platform: those terms are set per restaurant and per country, so a figure attached to a brand name would be wrong for most people reading it. Use the rate on your own invoice.
- What we checked against
- The platform claims quoted here were read on their own pages on 24 August 2026: Uber Eats’ merchant site and help centre, Bolt Food’s merchant page, and Glovo’s partner portal. Those pages change. Where a detail here and a detail there disagree, believe theirs.
- What we left out
- No client, no contract term and no commercial figure belonging to any restaurant or platform appears on this page.
- The checklist
- Ticking a box changes nothing but the counter. Nothing is stored, sent or remembered — reload the page and it is blank again.
We can just do this for you
Everything above is a few weeks of somebody's attention: chasing managers, filling in forms, renaming photos, writing item names that get found in search, building combos, setting up modifiers, testing what survives the bag. It is not hard work. It is fiddly work, and it competes with running a kitchen.
So we do it. We have direct contacts at the platforms and introduce you ourselves, we build both stores in the back office while the contracts move, and we hand your team a one-page guide for the device. Your side of it is the kitchen and the prices.
And we are paid a share of the growth we bring, measured against your own numbers. A brand-new store has no previous numbers, so on a launch that share is simply of what we build from zero — and if nothing gets built, there is no invoice.
Leave a request with your city and what you cook. We will tell you which platforms are worth opening where you are, what it will take, and roughly when the first orders land. Free, and nothing is signed on your behalf.