What should labour cost be for an F&B outlet in Singapore?

There’s no single right number. Labour cost as a share of revenue depends on your format, your price point and how much prep happens on site, so a kiosk and a full-service restaurant shouldn’t share a target. What works is to calculate labour % for each outlet, every week, measure it against a target band you’ve set for that format, and act when an outlet drifts outside it. The group average is the number most likely to mislead you.

Why labour cost needs closer watching in 2026

Pressure on wages isn’t easing. In SNEF’s Survey on Wage and Employment Outlook 2026/2027, rising manpower cost remained the top manpower challenge, flagged by 83% of employers, up from 79% in 2025 (fieldwork June–August 2026; 300+ companies across 20 industries).¹

For F&B specifically, the Restaurant Association of Singapore’s July 2026 Services Index found sales still 1.9% lower year-on-year, and noted that “manpower costs continue to rise via the Progressive Wage Model”, which operators absorb without matching revenue growth.²

When costs climb and sales don’t, a few points of labour % per outlet is the difference between a good month and a bad one.

How to calculate labour % per outlet

Labour % = total labour cost ÷ net sales × 100

Be consistent about what goes in each side:

Labour cost (for the same period as the sales):

• wages for full-time and part-time staff
• overtime
• employer CPF contributions
• foreign worker levies, if applicable
• allowances, bonuses and other staff benefits you pay

Net sales: revenue excluding GST. Decide whether you count delivery sales before or after platform commission, and stick to it. (Before commission is more common for this ratio. Commission is better tracked as its own line.)

Calculate it weekly, not monthly. A month blends a good week and a bad one into an average that’s too late to act on.

The outlet the group average hides

Here’s an illustrative week for a two-outlet group. These are made-up numbers, not a benchmark:

At group level, 30.5% looks fine. Outlet A is running six points above Outlet B. If the target band is 28–32%, Outlet A is the problem and the group number hides it.

That’s why every outlet needs its own line, every week.

How to set your own target band

1. Start from your own history. Pull 8–12 weeks of labour % per outlet. Your best normal weeks are a realistic target, not an aspirational one.
2. Set a band, not a number. A band (for example 28–32%) gives you room for a slow week without a false alarm, and a clear trigger when you’re outside it.
3. Separate formats. If one outlet is a kiosk and another is sit-down, give them different bands.
4. Look inside the day. An outlet can hit its weekly band and still be overstaffed every afternoon and short every lunch. Compare rostered hours with sales by hour.
5. Check rostered hours against paid hours. The roster is the plan. Payroll is what happened. The gap is often where the money goes.

What AI labour tools are reporting

Software that matches staffing to demand is a growing category. Operators using AI-driven labour tools have reported labour-cost reductions in published vendor case studies. Nory, a UK- and Ireland-based restaurant platform, publishes case studies reporting reductions from 7% (Rocksalt, Ireland) to 26% (Passyunk Avenue, London).³ ⁴

Read those for what they are: vendor-published results from a different market, from a product that also handles scheduling and payroll. They show the category’s potential, not what any one business, or any one tool, will achieve.

Where Backline fits

Backline calculates labour % for every outlet against a target band you set (default 28–32%), checks the roster against each outlet’s real sales peak, and flags gaps between scheduled and paid hours. When an outlet drifts, it drafts a roster adjustment, with a staff-hours target and a note for your ops manager, for you to approve. It doesn’t change rosters on its own. See how Staff & Roster works.