Restaurant Profit Margins in the UAE: What the Numbers Should Look Like

A restaurant can be busy, well-reviewed and still lose money. When an owner tells us “revenue is fine but there’s nothing left at the bottom,” they almost always have a margin problem, not a sales problem — and margin problems are found in the cost structure, not the takings.

This is the operator’s view of what a healthy UAE restaurant’s numbers should look like, where margin most often escapes, and the weekly discipline that keeps it from coming back. The benchmarks below are the published bands we run every diagnostic against — typical, indicative ranges, not advice; your real profit-and-loss statement is what gives the exact picture. Where an example uses figures, they are illustrative round numbers chosen so the arithmetic is easy to check.

What should a UAE restaurant’s cost structure look like?

Most of a restaurant’s profitability is decided by four cost lines, each best understood as a share of revenue. These are the bands behind the Restaurant Operating Index, and the same ones the Profit Leak Audit ranks your numbers against:

Cost line	Published band (% of revenue)	Ceiling	Where it goes wrong
Food cost	28–32%	32%	Supplier price creep, portioning, waste — untracked
Labour	25–30%	30%	Scheduled to comfort, not to covers
Prime cost (food + labour)	55–62%	62%	Each line defensible alone; the sum through the ceiling
Rent	6–12%	12%	A great concept in an over-priced unit
Delivery commission	3–6%	6%	Aggregator cut the dine-in margin can’t carry

Two notes on reading the table. Delivery commission is stated as a share of total revenue — the per-order commission rate on the app is a different number, and the bridge between the two is worked below. And prime cost is not a fifth line; it is food and labour added together, which is why its band is not the sum of the two ceilings. The prime-cost read explains why the sum, not either part, is the survival gauge.

Add controllable overheads and you can see why two restaurants with identical revenue can sit on opposite sides of profitability. The healthy one keeps each line inside its band and watches them weekly. The struggling one lets them drift and finds out at month-end, when it is too late to act.

What is actually left after the four lines?

Walk one illustrative month to the bottom. Say revenue is AED 300,000 (net of VAT — more on that below), and every line sits comfortably inside its band:

Line	Share	AED
Revenue	100%	300,000
Food cost	30%	90,000
Labour (fully loaded)	28%	84,000
Prime cost	58%	174,000
Rent	10%	30,000
Delivery commission	5%	15,000
Four lines together	73%	219,000
Left for everything else	27%	81,000

Check it: 90,000 + 84,000 = 174,000 (prime, 58%); add 30,000 and 15,000 and the four lines take 219,000, which is 73% of 300,000; what remains is 81,000, or 27%. That 27% has to carry utilities, marketing, repairs, licences, insurance, finance, depreciation and the owner’s return. It is enough — comfortably — which is what “healthy” means in practice: not a fat margin on any single line, but four lines each held inside their range so the remainder is real.

Now the same restaurant after a year of nobody reading the sheet. Food has crept to 36%, labour to 32%, delivery has grown as a share of the business so commission takes 8% of revenue, and rent is unchanged at 10%:

Line	Healthy	Drifted	Difference (AED)
Food cost	30% · 90,000	36% · 108,000	+18,000
Labour	28% · 84,000	32% · 96,000	+12,000
Rent	10% · 30,000	10% · 30,000	0
Delivery commission	5% · 15,000	8% · 24,000	+9,000
Four lines together	73% · 219,000	86% · 258,000	+39,000
Left for everything else	27% · 81,000	14% · 42,000	−39,000

Same revenue. Same room, same menu, same team. The remainder has halved — from 81,000 to 42,000 — and none of the four lines moved dramatically enough to alarm anyone on its own. That is the whole story of “revenue is fine but the profit is gone”: a few points on several lines at once, compounding quietly, discovered late.

What do VAT and corporate tax do to the margin?

Two lines that are not operating costs still decide what reaches you, and both are worth stating precisely because operators routinely misread them.

VAT is not revenue. The UAE levies VAT at 5% at the point of sale, and a business whose taxable supplies exceed AED 375,000 a year must register. The 5% you collect is remitted, never earned — so every percentage in this piece is a share of revenue net of VAT. Read your food cost against VAT-inclusive takings and you flatter yourself by five points of denominator; every band in the table above assumes you have not.

Corporate tax sits below the operating line. Taxable income is taxed at 0% up to AED 375,000 and 9% above that threshold. Illustratively, a restaurant with AED 600,000 of taxable income pays 9% on the 225,000 above the threshold — AED 20,250 — leaving 579,750. The operating margin is the number you control; the after-tax margin is the number you keep. State both, and let your accountant confirm the treatment for your entity — the rules carry conditions and thresholds this piece does not cover.

Food cost: where does the leak creep in?

Food cost rarely fails in one dramatic move. It creeps — a supplier price rise that was never renegotiated, portions that grew, waste that nobody logged. Because each change is small, it hides; because they compound, they hurt.

The gauge that catches it is the gap between theoretical and actual food cost. The costed recipes say the menu, at the mix you sell, should run at 29%; the inventory movement — opening stock plus purchases less closing stock — says it ran at 33 of every 100 dirhams. That gap is the most informative number in the kitchen: portioning, trim waste, unrecorded comps and staff meals, spoilage, or theft. Operators who track only the theoretical number are reading the menu, not the business.

The published ceiling is 32%. The structural red-line — the point at which our rescue check reads a line as urgent rather than drifting — is 38%, deliberately set above the ceiling: a lead is urgent when the model is structurally past the line, not merely at it. Between 32 and 38 is the zone where a weekly reader fixes things cheaply and a monthly reader discovers them expensively.

This is also the line where measured recovery is largest. In the one engagement we are cleared to name, Parco Group’s Jebel Ali operation was running food cost at 44%; over a 120-day reset of purchasing, portioning, menu pricing and waste control it came to 29% — the consented figures are on record. Every other engagement stays anonymised, but the mechanics are the same everywhere: costed recipes, standardised portions, a weekly food-cost number measured against a deliberate target, and a habit of renegotiating supply rather than accepting the invoice. (More in food-cost control and menu engineering.)

Labour: schedule to covers, not to comfort

Labour is the second-largest controllable line, and the most common failure is rostering to feel safe rather than to demand. The wage line then moves with nobody watching it.

Two disciplines hold it. The first is loading the number fully: in the GCC, labour is wages plus visas, medical, insurance, accommodation, transport and the end-of-service entitlement accruing quietly across every contract — count the payslip alone and the labour line you are managing to is fiction. The cost-per-head read walks that stack. The second is scheduling from the forecast: labour hours derived from forecast covers by daypart, priced fully loaded, checked each week as a share of sales against the 25–30% band — the method in how many staff a restaurant needs. The red-line here is 35%; past it, the roster is not a scheduling problem but a structural one.

If labour is your heaviest half of prime cost, the Labour Productivity read shows in two minutes whether the schedule is matched to covers or to habit.

Rent: the fixed cost you negotiate once

Rent is the one big line you cannot adjust after the fact, which is exactly why it has to be right before you sign. The published band is 6–12% of revenue, and the arithmetic runs backwards from the lease: a unit at AED 30,000 a month needs revenue of AED 250,000 a month to sit at the 12% ceiling (30,000 ÷ 0.12) and AED 500,000 to sit at the 6% floor (30,000 ÷ 0.06). If the feasibility says AED 180,000 a month, the rent is 16.7% before a single cover is served, and no amount of operational excellence fully rescues the model.

Once you are trading, the rent-to-revenue ratio is the only lever left — and it moves through the numerator (renegotiate, sublet, restructure) or the denominator (grow revenue into the space). The rent-versus-revenue check does the backwards arithmetic for any lease you are weighing, which is why the launch decisions around location and lease matter so much: it is the one line you cannot roster your way out of.

Delivery commission: how does a per-order rate become a share of revenue?

Delivery added revenue for almost everyone — and quietly compressed margin for many. The confusion starts with two different percentages wearing the same name.

The per-order commission is what the platform takes from each delivery order — as a teaching band, typically 15–30% of order value depending on platform, category, delivery mode and the extras you sign up for; when the Khaleej Times surveyed Dubai operators in 2020, the quoted range was 25–30%, “up to 35%” with everything stacked on top. The commission as a share of total revenue — the Index line, banded at 3–6% — is that rate multiplied by how much of your business runs through the apps:

Delivery share of revenue	at 15% commission	at 25%	at 30%
10%	1.5%	2.5%	3.0%
20%	3.0%	5.0%	6.0%
30%	4.5%	7.5%	9.0%
40%	6.0%	10.0%	12.0%

Read the table and the leak is obvious: a restaurant doing a fifth of its business through the apps at 25% sits at 5% of revenue — inside the band. Let delivery grow to 40% of revenue at the same rate and commission alone takes 10% of everything you sell, more than most operators pay in rent. The dish never changed price; the channel changed everything around it. That is why the rescue check reads a per-order commission above 25% as a structural line, and why margin has to be read by channel — a healthy dine-in room can subsidise a losing delivery screen for months without anyone noticing.

The work is to understand the true margin per delivery order (commission, packaging and promotion all come off before food cost), price or engineer the delivery menu accordingly, and decide deliberately how much volume you want through a third party versus your own ordering. The aggregator economics read walks one order to the last dirham; the Delivery Margin Recovery tool separates your delivery contribution from dine-in, commission included.

Why can two restaurants with the same revenue end up on opposite sides?

Because volume scales whatever structure it runs on. If each cover contributes, a busy month compounds the gain. If each cover quietly costs — a dish priced for the table sold through the app at a discount, a Friday-shaped roster deployed on a Tuesday, a unit whose rent needed twice the revenue — then busier is simply the rate at which the business loses. The full room and the empty account coexist far more often than the queue outside suggests.

The corollary matters for the owner who wants to “sell their way out”. If prime cost is through the ceiling, more covers scale the shortfall. The order of operations in every turnaround we run is structure first — the four lines back inside their bands — and growth second, because growth on a broken structure buys more of the same problem. Break-even tells you how many covers a day the current structure needs; if that number is above what the room can physically serve, the answer is not marketing.

The discipline that holds it: a weekly P&L

The single habit that separates restaurants that hold their margin from those that lose it is a weekly profit-and-loss rhythm — seen by the owner, not just compiled by the accountant at month-end. Monthly numbers tell you what happened after you can no longer change it. Weekly numbers let you catch a drifting line while there is still a month to fix it.

The cadence is short and unglamorous. Same morning every week: sales by day, food cost from the inventory movement (never purchases alone), labour hours at actual loaded rates with overtime flagged separately, the delivery statements reconciled to an effective take, and the four percentages written on one line beside last week’s and the band. On the illustrative AED 300,000 month above, a two-point food-cost drift is AED 6,000 a month — AED 1,500 a week — and a weekly reader meets it in week two, while a month-end reader meets it four weeks late with the habit embedded. The line-by-line P&L read sets out the sheet; for multi-outlet groups, that same discipline scaled across branches is the HO Control System.

Find your biggest leak first

If revenue looks fine and the profit has disappeared, the fastest way forward is to find which line is costing the most and fix that first. The Restaurant Profit Leak Audit takes five numbers — revenue, food cost, labour, rent and delivery commission — and ranks your top three likely leaks against the bands above with an estimated monthly impact, in about two minutes. It is free, computed on your device, and it is the honest place to start before a full, P&L-based turnaround that resets the structure and installs the weekly rhythm that keeps it inside the lines.