Key takeaways
- Write the schedule from a covers forecast, not from last week's schedule.
- Build a station grid first, then fill it with names; that keeps coverage and fairness separate decisions.
- Check labor cost and overtime before publishing; overtime under federal law starts after 40 hours in a workweek.
- Publish on the same day every week, and check whether your city or state requires advance notice.
Start from demand, not last week
The most common scheduling mistake is copying last week's schedule and changing names. That works until demand moves, and in restaurants demand moves a lot. US food services and drinking places sales were lowest in February 2025, at $75.8 billion, and highest in May, at $92.8 billion, a swing of more than a fifth in a few months[1]. Your own restaurant will have its own pattern by day, by meal period and by season.
Source: US Census Bureau, 2025[1]
Show the data
| Item | Value |
|---|---|
| February 2025 | 75.8 billion USD |
| May 2025 | 92.8 billion USD |
Labor is also one of your two largest costs. The National Restaurant Association estimates that food and labor each take roughly 33 cents of every dollar in sales[2]. A schedule that is a little too heavy every week adds up to a real loss over a year; one that is too light shows up in slow service, burnt-out staff and bad reviews. Writing from a forecast is how you find the middle.
The weekly scheduling routine
Forecast covers by meal period
Use the same week last year, the last four weeks' trend, the reservation book and known events. Write a covers number for each lunch and dinner, and split dinner into hourly blocks if your peak is sharp.
Turn covers into station needs
Use your own productivity ratios, such as covers per server per hour or tickets per line cook, to set how many people each station needs in each hour.
Draw the station grid
Lay out shifts by station and time, with no names yet: two servers from 11, a third from 12, a closing server until 10, and so on. Stagger start and end times to follow the curve.
Collect availability and requests
Use one channel for availability and time-off requests, with a clear cut-off, such as Wednesday noon for the schedule published Friday.
Fill the grid
Assign people to shifts, balancing skill, fairness, requested hours and cost. Put your strongest people on the busiest shifts and pair new hires with trainers.
Check cost and rules
Total the hours and wages against forecast sales, flag anyone heading toward overtime, and check rest periods, minors' hours and any local scheduling law.
Publish and lock
Publish on the same day each week, then handle changes through a written swap process approved by a manager.
Build the station grid before adding names
Separating coverage from people stops two problems. First, you will not under-staff a busy shift because your favorite closer asked for the night off. Second, you can see whether shifts follow demand: a grid where everyone starts at 4 p.m. and leaves at close is easy to write but expensive and tiring.
| Station | Tuesday dinner | Friday dinner | Notes |
|---|---|---|---|
| Host | 1 (4:30-9:30) | 2 (4:30-10:00, 5:30-9:00) | Second host covers phones and the waitlist at peak |
| Servers | 4 (two staggered starts) | 7 (three staggered starts) | First cut after the peak |
| Bartender | 1 | 2 | Second bartender doubles as service bar |
| Runner or busser | 0 | 2 | Friday only, 6:00-9:30 |
| Line cooks | 3 | 5 | Sauté and grill doubled on Friday |
| Dish | 1 | 2 | Second dishwasher from 6:00 |
Write the grid once for a typical week of each season and adjust it with the forecast. Most weeks you will change only a few cells, which makes the routine faster than it looks.
Check labor cost and overtime before you publish
Under the Fair Labor Standards Act, non-exempt employees must be paid at least one and one-half times their regular rate for hours over 40 in a workweek[3]. Many states add their own daily overtime or rest rules. Before you publish, total each person's scheduled hours and flag anyone above 36 or so, because a covered call-out or a late close will push them over.
- Scheduled labor cost as a share of forecast sales, by day and for the week.
- Each person's total hours, with overtime risks flagged.
- Minors' hours against federal and state limits.
- Rest time between a closing shift and the next opening shift.
- Breaks, where your state requires meal or rest periods.
- Trainers scheduled alongside trainees.
Employees in food services and drinking places averaged 25.1 hours a week in September 2026[4], so many of your staff are part-time and some will want more hours. Balancing those requests fairly, and explaining how hours are allocated, prevents a lot of resentment.
Publish on time, every time
A late schedule is one of the most common complaints in restaurant teams, because staff need to plan childcare, classes and second jobs. Pick a publishing day and stick to it. Two weeks ahead is a good standard, and in some places it is the law: Oregon's predictive scheduling law requires large employers in retail, hospitality and food services, those with at least 500 employees worldwide, to give written schedules at least 14 calendar days in advance[5]. Several cities have similar fair workweek rules, so check your local labor department.
Post the schedule in one place everyone can see on their phone, and make that place the official record. If the paper schedule in the office and the app disagree, staff will not know which to trust.
Shift swaps and changes
- Staff arrange swaps between themselves, with people trained for the same station.
- Every swap is requested in writing, in the scheduling app or on a swap form.
- A manager approves or declines within a set time and checks overtime and rest periods.
- The schedule is updated in the official place; unapproved swaps are not valid.
- Changes the restaurant makes after publishing are communicated directly to the people affected.
Track how often you change the schedule after publishing. Frequent changes usually mean the forecast or the availability process needs work, and in fair workweek cities they can also trigger premium pay.
Fairness rules your team can see
Most scheduling conflict comes from staff believing that good shifts go to favorites. Write down how shifts are allocated and share it, so decisions can be explained. The rules do not have to be complicated, but they should be the same for everyone and applied every week.
- Busy, high-earning shifts rotate among servers who meet the performance standard, rather than going to the same people every week.
- Requests submitted before the cut-off are honored where coverage allows, in the order received.
- Holiday shifts rotate year to year, with a record kept of who worked which holiday.
- Hours above an agreed minimum go first to those who asked for more hours and are trained for the station.
- New hires get at least one busy shift a week alongside a trainer, so they learn real service.
When you must break a rule, say why. A short message such as 'You are on Saturday again because Jamie is training on your section' prevents most complaints before they start.
Example: building a Friday from the forecast
Example (fictional): a 90-seat neighborhood restaurant forecasts 210 Friday dinner covers, with 60 between 7 and 8 p.m. The manager uses a ratio of about 12 covers per server per hour at peak, so the 7 p.m. hour needs five servers on the floor, plus two more to cover the shoulders and sections that turn late. She schedules three servers from 4:30, two from 5:30 and two from 6:30, with the first cut at 8:30.
In the same example, the chef doubles sauté and grill in the kitchen from 6 p.m. A runner works 6 to 9:30. The draft totals 7% more hours than last Friday, which matches the higher forecast. One server is at 38 hours for the week, so the manager moves her Sunday brunch shift to a part-timer who asked for more hours. The schedule is published at noon on Friday for the week starting the Monday after next.
Review the schedule against what happened
Each Monday, compare last week's scheduled hours with actual hours and actual covers. Note where you sent people home early, where you called people in and where service suffered. Those notes improve next week's forecast more than any rule of thumb. When the routine is stable, scheduling software can save time; our comparison of restaurant scheduling software and guide to labor forecasting cover the next steps.
Frequently asked questions
How far in advance should a restaurant schedule be posted?
Two weeks ahead is a good standard. Some states and cities require advance notice for large employers, such as Oregon's 14-day rule, so check local law.
What is the best way to forecast restaurant staffing?
Start from covers: the same week last year, recent trend, the reservation book and known events. Convert covers into station needs using your own productivity ratios.
How do I avoid overtime on a restaurant schedule?
Total each person's hours before publishing, flag anyone close to 40, and leave room for call-outs and late closes. Approve every swap with overtime in mind.
Should staff be allowed to swap shifts?
Yes, with people trained for the same station and with manager approval recorded in the scheduling app or on a swap form.
References
US Census Bureau (2025). Monthly Retail Trade: Estimates of Monthly Retail and Food Services Sales by Kind of Business (opens in a new tab) ↑
National Restaurant Association (2026). Restaurants remain resilient despite challenging business conditions (opens in a new tab) ↑
US Department of Labor, Wage and Hour Division (2026). Wages and the Fair Labor Standards Act (opens in a new tab) ↑
US Bureau of Labor Statistics (2026). Current Employment Statistics: average weekly hours, food services and drinking places (CES7072200002) (opens in a new tab) ↑
Oregon Bureau of Labor and Industries (2026). Predictive Scheduling (opens in a new tab) ↑