Building a weekly rota from the occupancy forecast
The rota should be a translation of expected demand into people and hours. Most are a translation of last week's rota. When a rota produces unusual overtime patterns, the pay method matters as much as the hours; this legal overview reviews one specific calculation approach.
Ask a manager how they built next week's schedule and the honest answer is usually that they opened the previous week, changed the names in a few cells and published it. This is not laziness. Building a rota from first principles takes hours, and copying forward is right often enough to feel safe. The Acas working-hours guidance offers practical background on working-time arrangements and limits.
It is right until demand moves. Then the schedule carries last week's shape into a week with a different shape, and the mismatch shows up as either idle payroll or a team that spent Thursday underwater.
Start from the arrival and departure curve
Occupancy percentage is the number everyone quotes and the least useful one for scheduling. Two properties at eighty percent can have completely different labour requirements: one with a stable long-stay guest base, one turning over every room. The workload lives in the movement, not the level.
What you actually want, for each day of the coming week, is three figures: rooms departing, rooms arriving, and rooms staying through. Departures drive the housekeeping deep clean, arrivals drive front desk and porterage, stay-throughs drive the lighter service. A day with forty departures and forty arrivals is a completely different day from one with five and five at the same occupancy.
Departures, arrivals and stay-throughs, by day. Every hospitality rota question becomes tractable once these are on the page, and almost none of them are tractable from occupancy percentage alone.
Convert to hours before converting to people
The common shortcut is to jump straight from volume to headcount — forty departures, so four housekeepers. This hides the assumption doing all the work, which is the minutes-per-room figure, and it makes the schedule impossible to argue with when it turns out to be wrong.
Do it in two steps instead. First, volume times standard time equals required hours. Forty departure cleans at thirty-five minutes is roughly twenty-three hours of housekeeping labour. Second, required hours divided by usable shift length equals headcount. If a housekeeper delivers six productive hours in an eight-hour shift once you account for briefings, breaks, trolley setup and travel between floors, twenty-three hours needs four people, not three.
The two-step version survives scrutiny. When someone asks why Thursday needs an extra person, there is an arithmetic answer rather than an instinct.
Layer the fixed points
Some hours are not negotiable regardless of volume. The desk has to be staffed from the first arrival to the last, whether that is two guests or twenty. Breakfast has to run at breakfast time. Fire panel checks happen when they happen. These fixed points form the floor of the schedule, and volume-driven staffing sits on top of the floor rather than replacing it.
Getting this backwards is the most common structural error in small-property rotas. A quiet Tuesday tempts you to cut the desk to one person, and then that person cannot leave the desk to help with a bag, take a delivery or deal with a maintenance call. The floor exists for a reason.
Build in the slack you will need anyway
A rota scheduled to exactly the forecast will fail, because forecasts are wrong and people are occasionally ill. The question is not whether to carry slack but where to put it, and there are three honest options.
- Roster slightly above requirement and accept the labour cost. Simple, expensive, and appropriate at peak season when a service failure costs more than an extra wage.
- Roster to requirement and keep a named, paid on-call. Cheaper, and it works only if the on-call person can actually reach the property quickly.
- Roster to requirement and hold a list of deferrable tasks. When the week goes smoothly, the deep clean of the storage room gets done. When it doesn't, that hour absorbs the shock.
The third option is the one most small properties use without naming it, which is a shame, because naming it turns an invisible buffer into a planning tool. A written list of deferrable work makes it obvious when you have spent your slack three days running and are now operating with none.
Publish early, and mean it
The scheduling horizon — how far ahead staff know their hours — is one of the strongest levers on retention available to a small operator, and it costs nothing but discipline. Two weeks of notice is workable. Four is genuinely valuable to people. One is a business that will lose anybody who has other options.
Publishing early does mean the forecast is less accurate at the moment of publication, and some adjustment will be needed. That trade is almost always worth taking. A schedule that changes twice with notice is easier to live with than a schedule that appears on Friday for the week starting Monday.