Why Cloud Kitchens Fail in Dubai (and How to Avoid It)
Most cloud kitchens in Dubai fail on unit economics, not food quality. Delivery apps typically take 25–35% per order, packaging can run 8–12% of revenue, and food cost takes another 28–32%. That leaves very little for rent, staff and marketing — and a brand that competes on discounts, owns no customer relationships and started with thin working capital can be busy every night and still lose money.
The good news: each of those failure points is visible before you open, and most are fixable.
Note: Cost ranges are industry estimates from our cloud kitchen margins guide and UAE operator reports.
Why do cloud kitchens fail?
1. The commission is taken from every single sale
A dine-in restaurant pays delivery commission only on its delivery orders. A cloud kitchen pays it on almost everything. At 25–35% per order, the apps take a larger share of each sale than the food itself costs.
Here is one AED 100 order:
| Line | AED |
|---|---|
| Order value | 100 |
| App commission (30%) | −30 |
| Food cost (30%) | −30 |
| Packaging (10%) | −10 |
| Left for rent, staff, marketing, profit | 30 |
Illustrative, using industry estimates. Run a 20% promotion on top of that and the order can lose money.
2. The menu looks like everyone else's
Open any delivery app in Dubai and scroll: dozens of burger, shawarma, pizza and bowl brands, many with near-identical menus. When a customer cannot tell brands apart, they choose on price and promotions — the one thing a cloud kitchen can least afford. A narrow, distinctive menu with a clear reason to choose it beats a broad, generic one.
3. The food doesn't travel
Some dishes are excellent at the table and poor after 30 minutes in a bag: fries, crisp items, delicate sauces, steak. Bad travel means bad ratings, and on delivery apps ratings decide visibility. Test every dish by ordering it to a real address before it goes on the menu.
4. The app owns the customer
When someone orders through an app, the app keeps the relationship. You usually cannot message them, reward them or bring them back directly, so every repeat order carries the commission again. A cloud kitchen with no direct customers is renting its entire business from the platform.
5. Too little working capital
Many operators budget for the launch and not for the months after it. Ratings take time to build, and early months often run at a loss. Kitchens that start with less than about six months of operating costs in reserve are forced into discounting to survive, which makes the economics worse.
6. Growth before profit
Adding brands, zones and promotions to grow order volume can hide the fact that each order loses money. More volume at a negative margin fails faster.
Is this only a Dubai problem?
No. The model has been tested hard worldwide. In the US, delivery-only operator Local Kitchens closed more than half its locations in October 2025 (QSR Pro). The operators still standing tend to share the same traits: tight unit economics, distinctive food, and a growing share of customers who order directly.
How do you avoid it?
- Model the unit economics first. Cost every dish including 30% commission and packaging; if a dish does not work at full commission, fix the price or drop it. Our guide to whether a restaurant business is profitable in Dubai has the tests.
- Engineer the menu for delivery. Fewer dishes, higher margins, food that travels. See menu engineering for UAE cloud kitchens.
- Build direct orders from week one. A WhatsApp number or menu link in every bag, and a small reason to order direct. Every order moved off the apps keeps the commission — our guide to WhatsApp ordering for Dubai restaurants shows the setup.
- Keep a customer list. Every direct customer is someone you can bring back without paying the app again.
- Protect your rating. Fix travel problems fast and respond to every review; ratings drive app visibility.
- Start small. A licensed shared kitchen keeps fixed costs low while you prove the brand — see our cloud kitchen licence guide.
- Hold six months of working capital so you never have to discount your way out of a slow month.
What are the warning signs your cloud kitchen is in trouble?
- Order volume is rising but cash in the bank is not. A sign that orders lose money after commission and promotions.
- More than a third of orders need a discount to happen. You are competing on price, not on a reason to choose you.
- Your rating is slipping below the brands around you. Visibility on the apps follows.
- Almost no customer orders directly. Every repeat sale still pays commission.
- You are using working capital to cover monthly losses with no date by which that stops.
Any one of these is fixable. Several together mean it is time to change the menu, the pricing or the channel mix before adding anything new.
The pattern behind every failure
Nearly every failed cloud kitchen made the same trade: it let the delivery app own both the customer and the margin. The kitchens that last treat the apps as a way to be discovered, then build a direct relationship with every customer who comes back.