Scheduling so labor doesn’t eat your margin

Labor is the cost you control in real time — and the one that quietly sinks otherwise-healthy restaurants. Overstaff a slow Tuesday and you’ve given away the night’s profit; understaff a rush and you’ve traded reviews and burnout for a few saved dollars. The fix isn’t working your team harder. It’s scheduling to reality and watching one number.

Schedule to demand, not to habit

Most schedules are copied from last week, which was copied from the week before. Instead, build from what actually happens: pull the last 4–6 weeks of sales by day and by hour. You’ll see your real peaks and lulls — and they’re rarely evenly spread. Staff the peaks properly and trim the dead hours. A cook who starts at 4 instead of 3 on a slow day, times a month, is real money back.

Set a labor target as a percent of sales

“How many people do I need” is the wrong question. The right one is “what should labor cost as a share of the sales this shift will do?” Pick a target labor % for each daypart, forecast the shift’s sales, and schedule to hit it. Now every scheduling decision has a yardstick instead of a gut feeling.

Catch overtime before it happens, not at payroll

Overtime discovered on the payroll run is money already gone. The habit that prevents it: before you publish, add up each person’s scheduled hours for the week and flag anyone near the overtime line. Mid-week, glance at who’s trending over. Almost all overtime is a scheduling oversight, not a staffing necessity — it’s preventable if you see it a few days early.

Publish early and make swaps easy

A schedule posted the night before guarantees call-outs and no-shows. Publish at least a week out so people can plan their lives — you’ll get better attendance and less turnover. And give staff a clean way to swap shifts with approval, so coverage stays intact without a flurry of group-chat messages you have to referee.

Watch coverage against the floor, live

The plan is only half of it. On the day, keep an eye on who’s actually clocked in versus scheduled. A no-call during a rush needs a fast response; three people clocked in on a dead afternoon needs an early cut. The managers who protect margin are the ones watching coverage in the moment, not reconstructing it later.