Are you overstaffing or understaffing? The rota maths that protects your covers and your margins

Stop building rotas from habitLink

“Saturday is always busy” is not forecasting. It is guesswork wearing an apron.

A rota is not a list of names. It is a financial decision. Every shift you publish commits money before the first customer walks through the door. Get it wrong and the cost lands in one of two places: idle staff hours that destroy your labour margin, or slow service, lost covers and unhappy guests that damage revenue.

Most operators make the same mistake. They build the rota from habit. “Saturday is always busy.” “Tuesday lunch is always quiet.” “Sarah usually works Friday.” That is not forecasting. It is guesswork wearing an apron.

The correct order is simple: forecast demand, convert demand into required labour, build shifts around the demand curve, monitor the live service, and compare scheduled hours, actual hours and revenue. Do those steps in the wrong order and you pay for the mistake all week.

Start with the numbers you already own. Pull the last six to twelve months of data. Your EPOS, booking system and rota software should provide most of this. If they do not, create a basic spreadsheet. Perfect data is not required. Useful data is.

• Covers by day and hour
• Sales by daypart
• Average spend per head
• Bookings and cancellations
• No-shows and walk-ins
• Ticket times
• Actual staff hours, overtime and agency hours
• Local events, holidays and unusual trading dates

Break your trading day into clear dayparts: breakfast, lunch, afternoon, dinner, late bar, and check-in and breakfast service for hotels and B&Bs. Then break the busiest dayparts into smaller blocks. A Saturday dinner from 5pm to 10pm is not one staffing requirement. It is five different operating conditions.

At 5pm, you may need two front-of-house staff. At 7pm, you may need five. At 9.30pm, you may need three again. A single flat rota ignores this curve. That is how operators pay five people to cover a quiet opening period and then leave the floor short during the rush.

Forecast covers before you forecast staffLink

A strong forecast is based on comparable days, not wishful thinking.

Use booked covers as your starting point. Then add a realistic walk-in estimate. For example, a lunch service might run 18 booked covers plus 6 expected walk-ins from 12pm to 1pm, 34 booked plus 10 walk-ins from 1pm to 2pm, 22 booked plus 8 walk-ins from 2pm to 3pm, and 8 booked plus 4 walk-ins from 3pm to 4pm. That gives you a lunch forecast of 110 covers.

Do not use your best-ever Saturday as the benchmark. Use a sensible trading average. A strong forecast is based on comparable days, not wishful thinking. Adjust it for known demand drivers.

• School holidays and bank holidays
• Weather
• Local matches and concerts
• Nearby weddings and theatre schedules
• Market days
• Hotel occupancy
• Paydays
• Seasonal tourism

A rainy Tuesday in January is not comparable to a sunny Tuesday in August. Treating them as identical produces a bad rota before the shift even begins.

For accommodation businesses, forecast room demand separately from food-and-beverage demand. A full house does not automatically mean a full restaurant. Guests may eat elsewhere. A quiet room night may still produce a strong breakfast service.

Set labour standards by roleLink

There is no universal number. Use your own operation as the baseline.

Now convert covers into labour requirements. Set a practical standard for each department: front-of-house minutes per cover, kitchen minutes per cover, bar staff per expected transaction volume, hosts per booking load, housekeeping hours per occupied room, breakfast staff per expected breakfast cover, and managerial coverage per service.

There is no universal number. Your layout, menu, service style and team experience matter. A small pub serving a compact menu needs a different model from a large restaurant with table service and a complex kitchen. Use your own operation as the baseline.

Suppose dinner is forecast at 90 covers. Your front-of-house standard is six minutes per cover. 90 covers × 6 minutes = 540 minutes. 540 minutes ÷ 60 = 9 FOH hours. If the service runs for four hours, you need approximately 2.25 front-of-house staff on average. That does not mean scheduling 2.25 people. It means you might schedule one opener for setup, three staff during the peak and two staff for the close.

The average is useful. The shape of the service is what determines the actual rota.

Build shifts around the demand curveLink

The same start and finish time for everyone is convenient for the rota writer and expensive for the business.

Do not give everyone the same start and finish time. Use staggered shifts. A dinner rota might put a supervisor on 3pm–10pm, a prep and early-service team member on 4pm–9pm, two peak FOH staff on 5pm–11pm, additional peak support on 6pm–10pm, and a closer on 7pm–11pm.

This puts the most people on during the most valuable service window. It also removes people as demand drops. Use split shifts where they make operational sense. A team member working 11am–2pm and 5pm–10pm may be more cost-effective than paying them through a dead afternoon.

But do not use split shifts as an excuse for poor planning. They must be fair, compliant and workable for the employee. A rota that protects margin by exhausting the team will create absence, turnover and recruitment costs later.

Check UK working-time requirements before publishing shifts. Staff generally need 11 hours’ rest between working days, proper rest breaks and weekly rest. Legal compliance is the floor. It is not the strategy.

• Check the rules in the UK Government guidance on rest breaks and working hours

Price idle hours in poundsLink

Percentages hide the damage. Convert them into money.

Assume you schedule four staff for an extra three hours on a quiet Tuesday. The fully loaded average cost is £13 per hour. 4 staff × 3 hours × £13 = £156. That is £156 wasted in one day.

The same problem appears in smaller chunks: thirty minutes left on after close, one extra person starting an hour too early, staff waiting between lunch and dinner, overtime caused by weak handovers, and managers covering gaps because the rota does not reflect the booking curve.

An apparently harmless 30-minute overrun across six team members, five days a week, can become hundreds of paid hours over a year. Track scheduled hours against actual clocked hours. Otherwise, your rota is fiction.

Protect service during the peakLink

Cutting labour is not the goal. Protecting contribution is the goal.

Understaffing creates a different bill: longer ticket times, fewer table turns, more refunds and remakes, lower spend per head, poor reviews, staff staying late, managers pulled away from revenue-producing work, and guests who do not return.

If three extra staff cost £180 but prevent the loss of ten £45 covers, the decision is obvious. The wage bill is £180. The protected revenue is £450. That is not overspending. That is capacity protection.

Your rota must therefore include a minimum safe and serviceable staffing level. Never staff below the level required to run the kitchen, bar, floor and close safely. The cheapest possible rota is often the most expensive one.

Use a 30/30/30 service modelLink

This structure is simple. That is why it works.

For each main service, divide the operating period into three phases.

Prepare the first 30%: this is setup, prep and early arrivals. Keep the floor lean. Put the right kitchen and supervisory staff in place. Do not bring the full peak team in before there is work for them.

Cover the middle 30%: this is the peak. Schedule your strongest team here. Add the extra server, kitchen section or bar support. The busiest period is where speed protects both covers and reviews.

Release the final 30%: demand begins to fall. Release staff in stages. Keep the people needed for cleaning, close-down and late bookings. Do not carry the entire peak team until the doors are locked.

Measure the rota every weekLink

A strong Saturday can conceal a disastrous Monday-to-Thursday rota.

Track these figures by daypart: forecast covers, actual covers, scheduled labour hours, actual labour hours, labour cost, labour cost as a percentage of sales, average spend per head, ticket times, overtime, and complaints and refunds.

Do not judge the whole week as one block. A strong Saturday can conceal a disastrous Monday-to-Thursday rota. Look for patterns: are Monday lunches overstaffed every week? Is Saturday dinner regularly short? Are managers staying late because closing shifts are too thin? Are staff clocking in early without approval? Does poor service appear whenever bookings exceed a certain level? Are you paying for capacity that never sells?

Set a review point after every major service. Adjust the next rota while the information is fresh.

Turn demand into bookingsLink

A better rota only works when demand is visible.

Your website, Google Business Profile and booking journey must show customers when you have capacity. If your quieter dayparts are invisible online, the problem is not just staffing. It is demand generation.

A clear, mobile-first booking path matters: visible booking buttons, accurate opening hours, current menus and a strong Google presence.

Paid campaigns should also match your capacity. Do not spend heavily to fill a Saturday you already sell out while leaving Tuesday lunch empty. Websites and landing pages should be built around measurable enquiries, not decoration.

The same measurement discipline applies outside hospitality: identify the leak, explain the cost and fix the point where revenue disappears.

• For a pub-focused example of a simple booking and enquiry approach, see PubLandlord
• For conversion-focused website work, see JetAds
• For a plain-English audit approach used by UK trade businesses, see VU1

Follow the correct orderLink

That is rota management. Not copying last week.

Do not start by cutting staff. Start by measuring demand. Then follow ten steps: forecast covers by daypart, establish labour standards by role, map required staffing against each service phase, build staggered shifts, set minimum service levels, publish the rota within legal and practical constraints, track actual hours against the plan, adjust the next rota using evidence, market the empty capacity, and recalculate the margin.

That is rota management. Not copying last week. Not giving everyone the same hours. Not celebrating a low wage percentage while guests wait 45 minutes for food.

• If you want a deeper review of your demand, booking flow, staffing assumptions and margin leaks, start with our free revenue and marketing audit