"We spent €13,000 on wages last week" is a number without a meaning. It could be excellent or catastrophic depending on what came through the door.
Labour cost percentage - wage cost divided by revenue for the same period - is the version with meaning, and it is the single metric most worth building a weekly habit around.
Get the numerator right
The most common mistake is using base pay only. The number you want is total cost of employment for the period:
- Base wages for hours actually worked
- Overtime and any shift premiums
- Employer social security contributions
- Pension contributions
- Accrued holiday pay
- Agency or contractor spend covering the same work
That last one matters more than people expect. A business that looks efficient on payroll while spending heavily on agency cover is not efficient; it has moved the cost to a different line.
Depending on the jurisdiction, employer on-costs typically add 15–30% on top of gross pay. A percentage calculated on base pay alone is not wrong by a rounding error - it is wrong by roughly a fifth.
Get the denominator consistent
Pick net revenue (excluding VAT) or gross and never change it. Compare like periods - a week against a week, aligned to the same day boundaries. If your revenue is recognised on a different cycle to your payroll, align them before you start, not after you have six months of data you cannot compare.
Know your target before you measure
Typical ranges vary hugely by sector:
| Sector | Typical labour cost % |
|---|---|
| Full-service restaurant | 28–35% |
| Quick service | 25–30% |
| Retail | 10–20% |
| Hotels | 30–40% |
| Care | 55–70% |
Treat these as orientation, not as targets. Your target depends on your rent, your margins and your model - a business with low occupancy costs can afford a higher labour percentage and still be more profitable than a competitor with a lower one.
What matters is that you choose a number deliberately, write it down, and measure against it. A target you inherited from an industry article is still better than no target, but only just.
Weekly, not monthly
Monthly is too slow. By the time a month-end figure tells you the percentage drifted, five weeks of rotas have already been built on the assumption it hadn't.
Weekly gives you a correction window. It also smooths less, which means you see the shape - the Tuesday that is always overstaffed, the week after a bank holiday where revenue drops but the rota didn't.
Read the movement, not the level
A single week's figure is noisy. What you are watching for is direction and cause:
Percentage up, revenue flat - you scheduled more hours for the same work. Look at the coverage requirement.
Percentage up, revenue down - the rota didn't respond to the drop. This is the most common and most correctable pattern: demand fell, hours didn't.
Percentage down, revenue up - good, but check the service metrics before celebrating. Understaffing shows up in the labour percentage as a win and in the customer numbers three months later as a loss.
Percentage down, revenue flat - genuine efficiency, or you have quietly moved hours into an unrecorded category. Check the scheduled-versus-actual gap.
Where the lever actually is
When the percentage is wrong, the reflex is to cut hours. It is rarely the best lever available.
Shift shape usually beats headcount. Moving a shift start by an hour to sit through the peak rather than beside it can remove more cost than cutting a person, without reducing coverage when it matters.
Overlap is the quietest waste. Two shifts spanning the same quiet hour is one paid hour of nothing, five days a week, fifty weeks a year.
Overtime normalisation is the most expensive. Regular overtime at a premium rate to cover a structural shortfall costs more than hiring for the shortfall. If the same four hours of overtime appear every week, it isn't overtime.
Skill coverage is the counterintuitive one. Understaffing on capability forces overstaffing on headcount, because you need three people present to guarantee one of them can handle the situation.
The five-minute weekly version
- Total cost of employment for the week, including on-costs
- Revenue for the same week
- Percentage, and the difference from your target
- Percentage last week, and the direction
- One sentence on why it moved
Five lines, once a week. It is the difference between managing labour cost and finding out about it.