Playbook · Restaurants and cafes 3 min read
Schedule your restaurant staff to the rush, not to last week.
Most schedules copy last week. Your sales and floor data already show when the rush hits. Build the schedule from that, and let staff phones handle sick calls and swaps.
This is how we'd set up scheduling for a restaurant. It's a playbook, not a report on a client.
Schedules built on habit
Most restaurant schedules start as a copy of last week. The manager adjusts for a holiday or a big booking and posts it. Then the rush lands at 6:15 instead of 7, two servers stand around at 3 p.m. and the line is a cook short on Saturday.
Both mistakes cost money. Too few people and tables wait, tickets stall and guests don't come back. Too many and you pay for it. In Manitoba, you owe staff who report for work and get sent home at least three hours of pay, or their full shift if it's shorter. Asking for volunteers to go home doesn't get around it, and Employment Standards has investigated restaurants over it.
- 4.7%
- job vacancy rate in accommodation and food services in August 2026, the highest of any sector, per Statistics Canada
- 87%
- of Canadian operators named labour costs as a pressure, per Restaurants Canada
- $16.40
- an hour, Manitoba's minimum wage since October 1, 2026, per Manitoba Employment Standards
Cutting hours is the obvious move, and it backfires when it leaves the rush short. Hours only save you so much. Guests served is the number with no ceiling. Put the hours where the guests are, and the same payroll serves more tables.
Why a scheduling app won't fix it
Scheduling apps are good at the paperwork. They post the schedule, collect availability and handle swap requests. They don't know when your guests show up, which station backs up or what a slow Tuesday in March looks like. So the manager still builds the schedule from memory, and the app sends it out.
The data to do better already exists. Your cameras count guests by the hour. Your POS and kitchen screen track your orders and show when the line falls behind. The schedule should come from that.
What we'd build
- Forecasts guests and orders by the hour from your POS, kitchen and floor data, plus reservations and the weather.
- Drafts a schedule for servers and cooks that matches the forecast, within each person's availability.
- Shows the labour cost of the draft against forecast sales before you post it.
- When someone calls in sick, texts the people who are free and trained for that station, and gives the shift to the first yes.
- Checks every swap request keeps the shift covered before approving it.
- Sends schedules, changes and reminders straight to staff phones.
- Real traffic
- Shifts that match
- Sick call covered
- Labour on budget
Want to see this mapped to your business?
The manager still approves the schedule. They just stop rebuilding it from scratch every week.
Not a restaurant?
Any shop with a rush can do this. A cafe can put a second barista on from 7:30 to 9 instead of all morning. A bakery can bring counter staff in for the lunch line instead of at open. A salon or a retail shop can find its dead hours and stop staffing them like Saturdays.
Start small
Pull last month's sales by hour and lay last month's schedule over it. Mark the hours with too many people and the hours with too few. Fix the worst ones first. Then we automate sick calls and swaps.
This is the last of three restaurant playbooks. The first two cover your cameras and your orders.
Want a schedule that follows your rush?
Bring last month's schedule and sales to a free 30-minute call. We'll show you where your hours and your rush don't line up.