Occupancy tells the hotel how full it will be, not how much work it will be. Here is why two days at the same occupancy can carry very different workloads, what the forecast has to contain before operations can staff from it, and a test you can run on last month's data.
Hotel Labor Forecasting Starts With Movements
A revenue manager forecasts room nights and revenue because pricing and the budget run on them. Operations staff against something else. Housekeeping cleans rooms, and a departure takes twice as long as a stayover. The front desk handles check-ins and checkouts. Breakfast serves guests, not rooms.
None of those numbers sits in an occupancy forecast. So each department builds its own. Housekeeping counts last year's departures, the restaurant guesses covers from the weekday, and the front office plans from the arrivals list two days out. Seven teams end up with seven versions of next week, inside the same hotel.
The revenue manager's forecast is usually the best of them. It is also the one the departments use least, because it stops one step short of what they need.
Same Occupancy, Different Workload
Take a 120-room city hotel. On both days, 102 rooms are occupied: 85 percent.
On Friday, the business guests leave and the weekend guests arrive. Housekeeping has 90 departures and 12 stayovers. On Tuesday, the midweek guests stay on. Housekeeping has 20 departures and 82 stayovers.
Use 30 minutes for a departure clean and 15 minutes for a stayover. Friday needs 2,700 minutes for departures and 180 for stayovers, about 48 hours in all. Tuesday needs 600 minutes for departures and 1,230 for stayovers, about 30.5 hours. Same occupancy, and Friday carries 57 percent more housekeeping work.
The rest of the hotel moves the same way. The front desk checks out 90 rooms and checks in 90 more on Friday, against 20 of each on Tuesday. Breakfast on Saturday morning serves leisure guests at close to two per room, while Wednesday morning serves business guests at little more than one per room. An occupancy forecast reads all of these days as identical.
What Rostering by Occupancy Costs
Now give housekeeping the occupancy number and nothing else. With 85 percent on both days, a fair roster puts the same team on each, at around 39 hours, the average of the two days.
On Tuesday, the team finishes almost nine hours of work early, and the hotel pays for hours it did not need. On Friday, the team falls almost nine hours short. Rooms are released late; arriving guests wait in the lobby, and the shortfall is covered by overtime or rushed cleans.
Repeat that once a week and the hotel pays for about 455 hours a year that it did not use, and buys roughly the same number again at overtime rates. Neither cost appears in the revenue report, so neither gets traced back to the forecast. Labor is the largest cost line in most hotels, and here it is planned against a number that cannot see the work.
What the Forecast Has to Contain
The fix is not a second forecast for operations. It is the same forecast, carried one step further. For every day ahead, and split by segment, it should show:
- Arrivals and departures, so housekeeping and the front desk see the movements.
- Stayover rooms, so the cleaning workload adds up.
- Guests in-house, so breakfast and the restaurant plan cover, not rooms.
Segment matters because segment drives both length of stay and guests per room. A weekend leisure guest arrives on Friday, stays two nights, and brings a partner. A project team arrives on Monday, stays four nights, and travels alone. Forecast the segments correctly, and the movements will follow.
None of this changes what the revenue manager decides. It changes what the forecast hands over. The staffing decisions stay with the department heads, who know their teams, their contracts and their productivity far better than any forecast does.
Deliver It When the Roster Is Set
Most hotels set the roster two to three weeks ahead. That puts the movement forecast inside the operational horizon, where the forecast is judged on accuracy, not on the actions it triggers. The three horizons, and how each one should be measured, are set out in Your 90-Day Hotel Forecast Missed. That Was the Point.
Two practical points follow. Save the forecast on the day the roster is built, and measure accuracy on departures and stayovers at that point, not only on room nights. A forecast that hits occupancy but misses departures by 20 rooms sends a housekeeping team home early or keeps it late, and the occupancy number never shows the miss.
Then agree a fixed day for the handover. Every Thursday, for example, the forecast for the weeks ahead goes to housekeeping, the front office and food and beverage, each reading its own lines from the same numbers. The hotel forecasting guide describes the weekly cycle the handover fits into.
Run the Test on Last Month
Pull last month's four busiest changeover days and four quietest ones. For each day, put three numbers side by side: occupancy, departures plus stayovers, and the housekeeping hours actually worked.
If the hours follow occupancy, housekeeping is rostering from the wrong number. If the hours follow the movements, someone in the hotel is already building the movement forecast on their own, usually in a spreadsheet nobody else sees. Either way, the revenue manager's forecast is not reaching the people who schedule, and the gap is one handover away from closing.
Thursday's post takes the same question to the general manager, who decides whether the forecast reaches every department in the building.
Demand Calendar is a total-revenue forecasting and profit system for hotels. It sits alongside your RMS and PMS, not instead of them.
A forecast that stops at occupancy tells the hotel how full it will be. A forecast that shows arrivals, departures, and guests indicates how much work it will have, and that is the number the hotel pays for every day.
BOOK A STRATEGY CALL → demandcalendar.com/book-a-call
