What prime cost is
Prime cost is the sum of your cost of goods sold — food and beverage — and your total labour cost, expressed as a percentage of revenue. Total labour means everything: salaries, wages, visa and medical costs, gratuity accrual, overtime, staff meals and accommodation where you provide it. Not just the payroll line.
It exists as a metric because these are the two cost blocks a restaurant operator can actually control day to day. Rent is fixed by a lease signed years ago. Utilities move within a narrow band. Marketing is discretionary but small. Food and labour are two of the largest cost blocks that operating teams can influence day to day, which is why they are useful to manage together rather than in isolation.
Why it matters more than food cost alone
Food cost on its own is a misleading metric because food and labour trade against each other. Buying pre-portioned protein raises food cost and lowers prep labour. Making sauces in house lowers food cost and raises labour. Both can be correct decisions. Looking at either number in isolation will make one of them look like a problem when it is actually a trade.
Prime cost captures the combined effect, which is why it is the number to manage. If prime cost is stable at 62 percent, the mix between food and labour is a legitimate operational choice. If prime cost has moved from 62 to 67, something real has changed and it is costing you five percent of revenue.
Scale it: on a site doing AED 500,000 a month, five points of prime cost drift is AED 25,000 a month — AED 300,000 a year, from one location. That is usually the difference between a site being worth keeping and worth closing.
Concept planning ranges
The ranges below are F&B Performance internal planning bands used as a sense-check in the diagnostic tool. They are not audited UAE market benchmarks. Your own approved budget, menu engineering, lease, service model and staffing plan should take priority when better evidence exists.
| Format | Food planning band | Labour planning band | Prime-cost planning band |
|---|---|---|---|
| Casual dining | 29–35% | 25–32% | 55–64% |
| QSR / fast casual | 28–33% | 22–28% | 52–60% |
| Café / bakery | 23–30% | 25–32% | 50–60% |
| Fine dining | 30–38% | 30–38% | 60–70% |
| Delivery-first / cloud kitchen | 28–35% | 20–28% | 50–60% before delivery commission |
Two classification issues matter in the UAE operating context. First, define whether labour includes visa, medical, gratuity accrual and accommodation, and use that definition consistently across periods and sites. Second, keep delivery commission visible as a separate channel cost; a delivery-heavy site can show an acceptable prime cost while channel contribution is weak.
Do not use one prime-cost ceiling as a universal pass/fail rule. The real test is whether the remaining contribution after prime cost can cover occupancy, delivery/channel costs and the rest of the operating base while still reaching the site margin target.
How to calculate it properly
The formula is simple; the discipline is in the inputs.
Prime cost % = (Food cost + Beverage cost + Total labour cost) ÷ Net sales
Where operators go wrong:
- Using purchases instead of usage. Food cost is opening inventory plus purchases minus closing inventory. Using purchases alone means prime cost swings wildly with delivery timing rather than with actual consumption. This requires a weekly inventory count of at least your high-value items.
- Net sales, not gross. Sales after discounts, comps and voids. Using gross sales understates prime cost by whatever your discount rate is.
- Excluding parts of labour. Gratuity accrual is a real monthly cost even though you pay it years later. Staff meals are food cost. Visa amortisation is labour.
- Netting delivery revenue. If you record aggregator settlements net of commission, your sales base is understated and your prime cost percentage is overstated. Record gross, treat commission as a separate cost line.
Theoretical versus actual: the number that finds the money
Prime cost tells you where you are. Theoretical versus actual variance tells you why.
Theoretical food cost is what your recipes say you should have spent, given what you sold. Take the product mix from the POS, multiply by the costed recipe for each item, and you get the food cost that should have occurred. Compare it to actual usage.
The gap is one of four things: waste, over-portioning, purchasing error, or shrinkage. All four are addressable, and none of them are visible in the headline food cost number.
The arithmetic is what makes the variance worth investigating. If a site with AED 6 million annual sales should run at 30% food cost but actual usage is 34%, the four-point gap is AED 240,000 a year before considering whether the difference is explained by mix, recipe changes, waste, purchasing or stock-control issues.
Set a materiality rule: define the variance level that triggers investigation for each concept and category. A fixed universal percentage can be misleading when product mix, count accuracy and commodity volatility differ.
The weekly routine
Prime cost measured monthly is a report. Measured weekly it becomes a control. The routine that works:
- Sunday: inventory count on high-value and high-movement items. You do not need a full count weekly — the top 20 percent of items by value usually explains 80 percent of variance.
- Monday: calculate actual food cost from the count, pull product mix from the POS, calculate theoretical, and compute the variance.
- Monday: labour percentage against the scheduled hours and against sales, by daypart.
- Tuesday: a fifteen-minute review with the head chef and the GM covering three numbers — prime cost, variance, and labour productivity — and one action per exception.
The meeting is the point. Numbers circulated without a standing conversation get ignored within a month.