Real-time restaurant data: turn a number into a decision

It is 8 pm and sales look lower than last Friday. Are there fewer guests, a later sitting or large bills that have not yet been paid? The same number can have several explanations. Current data is valuable because you can investigate while the evening is still unfolding.

Decide what you need to know now

During service, a few questions are often more useful than a screen full of metrics:

• How many orders are waiting to be completed?
• How do recorded sales compare with a similar evening?
• Which reservations are arriving next, and are table statuses correct?
• Is demand concentrated on one dish or kitchen station?

Order flow and table status help the team immediately. Sales and staffing analysis support the shift manager. Everyone does not need the same view.

Read the definition before drawing a conclusion

Check whether sales include VAT and whether the report counts placed, paid or closed orders. An open bill may be treated differently from a completed purchase. Tips, refunds and cancelled orders also need clear definitions.

Average spend per order is not automatically spend per guest. When a group splits payment, the report’s definitions of orders and guests need to remain consistent.

Real-time means ongoing updates, but information is only as reliable as its recording. A table never marked as departed or a missed clock-out can distort the picture. Check connectivity and the latest update when a figure looks implausible.

Compare the same part of the evening

Do not compare today’s sales up to 8 pm with the whole of last Friday. Use matching intervals and consider opening hours, booking patterns, events and menu changes.

In a hypothetical example, today’s recorded sales are SEK 24,000 and the comparison period’s are SEK 30,000, using the same definition. The difference is 20%. Before changing the plan, check whether larger groups will arrive later or more bills remain open.

One weak hour is thin evidence for a permanent price change. Historical records help establish whether the pattern repeats.

Link each deviation to a check

If the kitchen queue grows, speak to the kitchen before accepting more orders or promising shorter waits. If a dish sells unexpectedly well, physically check available portions. POS records show dishes sold, not automatically what remains in the fridge.

If the labour cost ratio looks high, check hours, cost assumptions and whether sales are comparable. Costs during a live shift are often estimates. Consider workload, skills and employment terms before changing staffing.

Use Vendion for a shared view

Vendion Analytics++ brings together analysis of sales, product mix, staffing and bookings from the parts of the system you use. AI Boss can help answer questions about available records. Check the period and source information even when the answer comes through AI.

Finish with a short review after the shift: what deviation did we see, what did we do, and what happened? This gives both the current operating view and the completed report a clear purpose.