The Restaurant Scheduling  Guide

By Rob Bruni, founder of OpsEdge Hospitality, twenty years running restaurants, bars, clubs and resorts. Updated November 2026.

Free printable: Weekly Labor Cost Control checklist (PDF)

Most restaurant schedules are last week's schedule with the names moved around. That is why labor cost drifts: the floor is staffed for the week you remember, not the week that is coming. Scheduling to a forecast means estimating sales by day and daypart first, converting that into the hours you can afford, and only then writing names into slots. This guide walks through all three steps with a worked example and the rules that keep the schedule from sliding back.

Why the schedule is the fastest cost lever you have

Food cost takes a re-cost and a menu change to move. Rent takes a lease renewal. Labor moves the week you decide it should, because the schedule gets written every week. Two points of labor on a $2 million restaurant is $40,000 a year, and most of it is in the first and last hour of shifts that were never matched to demand.

The schedule is also where service lives. Over-staffing is expensive, under-staffing is expensive in a different way, and both come from the same mistake: staffing to a guess. A forecast is a better guess, written down, that you can check against what actually happened and improve every week.

Step 1: Forecast sales by day and daypart

Every POS can export sales by day and by hour for the last eight to twelve weeks. That is all you need.

1. Pull the last six to eight weeks of net sales by day and by daypart (lunch, dinner, late night, or by hour if you prefer).
2. Average each day of the week. Tuesday lunch averages Tuesday lunch, Saturday dinner averages Saturday dinner. Throw out any week with a holiday, a closure or a storm.
3. Adjust for what you know. A private party booked for Thursday, a street festival on Saturday, a forecast of rain on Sunday, a competitor opening. Write the adjustment and the reason next to the number so you can see later whether you were right.
4. Write the forecast for next week by day and daypart. Round to the nearest $100. Precision beyond that is pretend.

Eight weeks of history gets most operations within 10 percent of actual on a normal week. That is close enough to schedule to, and it gets better every week you compare forecast to actual.

Step 2: Turn the forecast into labor hours

Target labor percent is your hourly labor goal for that daypart (not total labor; salaried managers are scheduled separately). Loaded hourly rate is the average wage plus payroll taxes and benefits, usually wage × 1.15 to 1.25. Do this separately for front of house and back of house, because the rates and the targets differ.

A second way to get the same answer is sales per labor hour (SPLH): forecast sales ÷ SPLH target = hours. If your operation does $55 of sales per back-of-house hour on a good week, a $6,000 dinner forecast gives you 109 kitchen hours. SPLH is easier for managers to feel than a percent, and it transfers between weeks of different volume.

Whichever you use, the output is a number of hours per daypart per department. That number is the budget. The schedule has to fit inside it.

Step 3: Build the schedule

1. Start with the minimum crew. Every daypart has a floor below which you cannot open: one cook, one server, one bartender, a manager. Put those on first. They are fixed.
2. Add hours where the forecast says the sales are. Layer the second cook and the third server onto the peak hours, not onto the whole shift. A 4 p.m. to close server on a night that does not get busy until 6:30 is two and a half hours of labor for nothing.
3. Stagger starts and ends. The first and last hour of a shift are where labor leaks. Three servers starting at 4:00, 4:30 and 5:00 cover the same peak as three at 4:00 and cost an hour and a half less.
4. Check the total against the budget for each daypart. If it is over, cut from the shoulders before the peak.
5. Check each person against 40 hours. Overtime that is on the schedule is a choice. Overtime that appears on payday is a failure.
6. Then assign names. Skill, availability and fairness come in here, inside the hours you already decided you can afford.

A worked example

That is $350 on one night from stagger alone, with the same number of people on the floor at 7:30 when the party sits. Across a week it is usually $1,000 to $2,000 for a restaurant this size, which is two to three points of labor.

Rules that keep the schedule honest

• Post it the same day every week, at least a week ahead. Staff plan their lives around it, and late schedules cost you people.
• Clock-in window. No clock-in more than five minutes early without a manager override. Fifteen minutes early across 30 shifts a day is 7.5 unplanned hours.
• Cuts are decided by the manager, by the forecast, not by who asks. Set the cut order before the shift. "First cut at $X in sales" is a number, not a mood.
• Shift swaps need approval, and the approval checks overtime. A swap that puts someone at 44 hours is not a swap, it is a raise.
• Managers covering line positions go on the sheet. If the chef is on the line every Friday, the kitchen is short a cook and the labor number is hiding it.
• Compare forecast to actual every Monday. Which days were off by more than 15 percent, and why? The forecast only gets better if you check it.

The weekly rhythm

The whole routine is about 90 minutes a week once the first two weeks are built. Our Employee Schedule and Labor Cost Tracker does the forecast-to-hours math and flags overtime as you build, and the Labor Forecast and Staffing Guide covers the daypart targets by concept. The labor cost guide goes deeper on what belongs in the labor number.

Related guides and tools

Free tool: the sales per labor hour calculator turns next week's forecast into an hours budget for front and back of house. No signup.

FAQ