Most restaurant schedules are built the same way: last week's schedule, copied, with a few names swapped around. That works until a new menu launches, the patio opens, or a regular big table books every Thursday. Then the copy is wrong in ways nobody notices until a server is in the weeds at 7:15pm.
Here is the method I used for years as a GM, and the one we built Shiftnest around. It takes about an hour the first time and 20 minutes a week after that.
Step 1: Start from the forecast, not from the staff list
Before you open a spreadsheet, write down what you expect to sell, day by day and ideally by daypart. Use:
- The same week last year (adjusted for growth)
- The last four weeks of the same weekday
- Anything you know is coming: reservations, events nearby, a holiday, a local football game, forecast rain on a patio night
You do not need to be precise. A forecast within 10% is good enough to staff from. The point is that every shift on the schedule should exist because of a number, not because "we always have four servers on Friday".
Step 2: Convert sales into labor hours
Pick a target sales per labor hour (SPLH) for each section. If your Friday dinner forecast is $6,000 and you want front of house to run at $90 SPLH, you need about 67 FOH hours across that service. Divide by shift length and you have a headcount.
A simpler approach for smaller places is covers per server. If each server can handle about 20 covers in a three-hour peak, and you expect 110 covers, you need five or six servers at the peak and fewer either side of it.
Step 3: Build a staffing guide (once)
A staffing guide is a one-page table that says how many people you need in each role at each sales level. Build it once and you never have to think about headcount again; you just read the row that matches your forecast.
| Forecast dinner sales | Servers | Bartenders | Hosts | Line cooks | Dish |
|---|---|---|---|---|---|
| Under $3,000 | 3 | 1 | 1 | 2 | 1 |
| $3,000 to $4,500 | 4 | 1 | 1 | 3 | 1 |
| $4,500 to $6,000 | 5 | 2 | 1 | 3 | 1 |
| $6,000 to $8,000 | 6 | 2 | 2 | 4 | 2 |
| Over $8,000 | 7 | 3 | 2 | 5 | 2 |
Adjust it after every month. If you keep running short on bartenders at $5,000 nights, change the row.
Step 4: Collect availability before you assign anyone
Get availability in writing, and set a deadline. A good rule is that availability and time-off requests for a given week are due 14 days before it starts. That is also the notice period many predictive scheduling laws require, so it is a habit worth building early. (If you operate in Seattle, Chicago, New York City, Oregon or a handful of other places, read our guide to predictive scheduling laws before you go further.)
Keep three things on file for every employee:
- Days and times they can work
- Minimum and maximum hours they want
- Roles they are trained for
Step 5: Fill the anchor shifts first
Start with the shifts that make or break the week: Friday and Saturday dinner, Sunday brunch, any big event. Put your strongest people there first, including a clear shift lead for every service. Then fill outward toward the quieter shifts.
This order matters. If you schedule Monday first, you will run out of your best people before you get to Saturday.
Step 6: Stagger starts and cuts
Build every peak as a ramp. For a dinner rush from 6:30pm to 8:30pm:
- First server in at 4:30pm for side work and early tables
- Second and third at 5:00pm
- Fourth and fifth at 5:30pm
- First cut at 8:30pm, second at 9:00pm, closer stays
That pattern usually takes 8 to 12 hours a week out of a mid-size restaurant's FOH schedule with no change in service. Ten hours at $15 loaded is $150 a week, or $7,800 a year.
Step 7: Check it against the rules
Before you publish, look for:
- Overtime: anyone scheduled over 40 hours (or over 8 hours in a day in California)
- Clopens: a closing shift followed by an opening shift with less than 10 hours in between
- Minors: hour limits on school nights for staff under 18
- Breaks: required meal and rest breaks in your state
- Your labor target: projected labor cost as a percentage of forecast sales (see restaurant labor cost percentage for benchmarks)
Step 8: Publish early and in one place
Post the schedule at least a week ahead, two if you can. Put it in one place everybody checks: an app, not a photo of a whiteboard in a group chat. When the schedule lives in five places, staff work from whichever version they saw first.
Step 9: Handle changes with a swap policy
Shifts will change. What you want is for changes to happen in a way you can see and approve, rather than through private texts. A written shift swapping policy plus a tool that lets staff offer and pick up shifts themselves (with manager approval) removes most of the Saturday morning panic calls.
Step 10: Review the week
Every Monday, compare three numbers for the week that just ended: forecast sales vs actual, scheduled hours vs actual hours, and labor percentage. If actual hours were consistently above scheduled, you have a clock-in problem or a cut problem. If sales were consistently off, fix the forecast, not the people.
A quick checklist
- Forecast written down by day and daypart
- Staffing guide applied to the forecast
- Availability and time off collected 14 days out
- Anchor shifts filled first with a lead on each
- Starts and cuts staggered
- Overtime, clopens, minors and breaks checked
- Published 7 to 14 days ahead in one place
- Swap policy in writing
FAQ
How far in advance should a restaurant schedule be posted?
At least 7 days, and 14 days if you can manage it. In cities and states with predictive scheduling laws, 14 days is the legal minimum for covered employers.
How many hours should a restaurant manager spend on the schedule?
With a staffing guide and availability on file, 20 to 30 minutes a week. If it is taking you three hours, the process is the problem, not you.
Should I use a template or software?
A spreadsheet template is fine for one location under 15 staff. Once you have swaps, multiple roles and payroll to reconcile, software pays for itself in the hours you stop spending on it.
Build next week's schedule in Shiftnest
Shiftnest gives you a drag-and-drop rota builder with your sales forecast and labor percentage beside it, flags overtime and clopens before you publish, and sends the schedule to every phone. See the scheduling features or start free for up to 10 staff.
