Most restaurant schedules are built on gut feel and guilt. Here's a system for scheduling based on actual sales data so you stop over-staffing slow nights and burning out your best people.
The schedule is one of the most time-consuming things a restaurant owner or manager does every week. It's also one of the most consequential. A bad schedule either costs you money in over-staffing or costs you guests in under-staffing. Most operators get it wrong in the same direction: too many people on slow nights, not enough margin left over for anything else.
The most common scheduling mistake is building the schedule based on how last week felt rather than what last week actually did. "It was slow Tuesday" is not data. "$1,100 in sales on Tuesday between 5–10pm with 3 servers on the floor" is data.
Pull three months of daily sales from your POS, broken down by daypart if possible. What does your average Monday lunch do? Your average Friday dinner? Your average Sunday brunch? Build a sales forecast for the coming week based on those averages adjusted for any known variables - a local event, a holiday, a catering pickup.
Then schedule to the forecast, not to the hope.
A useful rule of thumb: you want to generate $30–$40 in sales per labor hour worked. If you're doing $1,200 in a dinner shift and you have 6 people working 5 hours each, that's 30 labor hours and $40/hour - healthy. If you have 8 people working the same shift, you're at $1,200 ÷ 40 hours = $30/hour - you're at your floor.
This isn't a hard rule - volume, concept, and service style all affect the number - but it's a quick gut check. If you're below $25/labor hour consistently, your scheduling is your biggest labor problem.
Start with your non-negotiables: the people who have to be there regardless of volume. That's usually your opener, your closer, your kitchen lead, and whoever is running the floor. These are your fixed labor costs for the shift.
Then add variable staff based on your forecast. If you're projecting a $2,000 Friday dinner, you know from experience that takes X servers and Y kitchen staff. If it comes in at $1,400, you're over-staffed - that's your cue to cut someone early, which requires having a conversation before the shift, not in the weeds at 7pm.
Cutting someone two hours into a slow shift saves you two hours of labor cost plus the employer-side taxes on those wages. Over a year of slow Tuesdays, that adds up to real money. Most operators don't cut early because it's uncomfortable - the employee is disappointed, the floor feels thin, nobody wants to be the bad guy.
The fix is to set expectations upfront. When you hire someone, explain that slow shifts happen and early cuts are part of the job. Build a culture where people understand it's the business being smart, not the manager being cheap. Employees who've worked in well-run restaurants expect this.
The more people you have who can do more than one job, the more scheduling flexibility you have. A server who can also run food, a line cook who can prep, a host who can bus - these people let you staff leaner without the floor falling apart. Cross-training takes time upfront and pays off for years.
One-week scheduling puts you in constant reactive mode. Two weeks out gives your staff enough notice to manage their lives, reduces last-minute callouts, and gives you time to find coverage before the day of. It also forces you to think about your business further ahead, which is a habit worth building.
Stage-aware courses, real calculators, and weekly check-ins — for less than one shift of labor a month.
Get started free