Key points
- Count people and arrival windows, not only occupied rooms.
- Separate confirmed attendance, assumptions and actual covers.
- Agree an equipment fallback before service begins.
Build the forecast with front desk
Front desk provides a dated snapshot of guests in house, known early departures, relevant arrivals and booked breakfast packages. The breakfast supervisor checks which entries describe entitlement and which indicate an intention to attend. A package that includes breakfast does not guarantee that the guest will visit the restaurant.
For groups, obtain the requested time, announced headcount and a contact able to confirm changes. Avoid named lists when a group reference and number will do. Specific dietary requests follow the property’s appropriate handling process; the forecast should not become a widely shared record of personal information.
Use a worksheet for each service window
| Field | Fictional example | Owner |
|---|---|---|
| Forecast version | Previous day, 6 p.m.; updated at 6 a.m. | Front desk |
| Eligible individual guests | 80 people | Front desk |
| Attendance assumption | 60–75%, central assumption 70% | Breakfast supervisor |
| Confirmed group | 24 people at 7.30 a.m. | Group contact |
| Other expected covers | 8 estimated walk-in purchases | Breakfast supervisor |
| Equipment constraint | One coffee machine unavailable | Engineering and restaurant |
Here, the central forecast is 80 × 70% + 24 + 8 = 88 people. The range is 80 × 60% + 24 + 8 = 80 to 80 × 75% + 24 + 8 = 92 people. These percentages are invented for the calculation. Each property should establish its own assumptions from relevant local observations.
Allocate the forecast to service windows
This fictional distribution uses the same totals of 80, 88 and 92 people. The 24 group members are included in the first window; all other attendance, including the 8 estimated walk-in purchases, is distributed across the three windows. Replace this allocation with the property’s own observations and confirmed timings.
| Illustrative window | Low | Central | High | Preparation decision | Owner |
|---|---|---|---|---|---|
| 7–8 a.m. | 30 | 34 | 36 | Prepare to welcome 24 people at 7.30 a.m. and an appropriate seating area | Restaurant supervisor |
| 8–9 a.m. | 30 | 34 | 36 | Plan additional clearing support if the peak is confirmed | Breakfast supervisor |
| 9–10 a.m. | 20 | 20 | 20 | Adjust replenishment progressively to observed attendance | Kitchen contact |
| Total people | 80 | 88 | 92 | Review the allocation at the next update | Front desk and restaurant |
Translate attendance into preparation decisions
A total is not enough: 24 simultaneous arrivals can create a peak even when the later service is quiet. Allocate expected attendance to time windows, then decide seating preparation, greeting roles, clearing rounds and replenishment arrangements. Separate work that can happen in advance from decisions that should wait for better confirmation.
The kitchen determines quantities and preparation methods through its own procedures. The operational worksheet communicates assumptions and changes; it does not replace food safety rules. The FDA Food Code is a model for jurisdictions, rather than a universal hotel operating procedure.
Include equipment limitations and fallback service
Before opening, the team checks which essential equipment is available and who receives incident reports. For the coffee machine in the example, food and beverage prepares an approved alternative, adjusts the queue arrangement and tells front desk what can be offered. A qualified technician decides technical action; restaurant staff should not improvise a repair.
The fallback identifies the service maintained, estimated capacity, accountable owner and next review point. If no acceptable alternative is available, management decides the offer and guest communication. An open engineering ticket does not itself explain how breakfast will operate. The restaurant supervisor confirms that colleagues understand the arrangement before guests arrive.
Close the forecast against actual attendance
At the end of service, record actual covers using the same unit as the forecast. Identify a late group, a changed booking or an incorrect walk-in assumption separately. Do not interpret a revised demand assumption as a kitchen performance problem when the information changed after preparation began.
Review the service windows that caused an observable issue: waiting, seating, replenishment or equipment capacity. Front desk and food and beverage select a specific adjustment for the next service. The RoomChecking solution for food and beverage teams provides a relevant starting point for discussing coordination needs. Keep the proposed worksheet’s ownership and definitions clear regardless of the tools chosen.
Frequently asked questions
Should the forecast change after every arrival?
Only when that change can still affect preparation. Set update points and an exception channel for material changes to avoid unnecessary competing versions.
Which attendance percentage should we use?
There is no universal rate. Use comparable local observations, retain a range and keep confirmed group attendance separate from uncertain intentions.