How to Forecast Restaurant Demand So Staffing and Prep Match the Rush

Labor and food are the two biggest costs a restaurant controls, and both are mostly decided before the shift begins. The schedule is written days ahead, the prep list the morning before. If the guess behind them is wrong, the shift is either overstaffed and overprepped, or short-handed and running out of the best-selling items.

Many operators still make that guess from memory: "Saturdays are busy, Tuesdays are slow." That works until a local event, a heat wave or a delivery promotion changes the pattern. This guide shows a simple, repeatable way to forecast demand by hour, turn it into staff schedules and prep quantities, and measure whether the forecast is getting better.

Step one: forecast by hour, not by day

A daily total hides the shape of the day. Two Saturdays with the same sales can need very different crews if one peaks at noon and the other at 7 p.m. Forecast covers, or orders for counter service and delivery, for every hour you are open.

Here is a worked example with hypothetical numbers for one Friday dinner:

The base is the average for the same weekday and hour over the last four to six weeks. Adjustments come from things you know: events, holidays, weather, promotions and school calendars. Keep a log of every adjustment and how it turned out, and your adjustments get more accurate over time.

Step two: turn demand into labor

Decide how many covers or orders each position can handle per hour at your standard of service. Then divide.

If a server comfortably handles 18 covers an hour, the 7 p.m. forecast of 104 covers needs about six servers on the floor. Do the same for the line, the bar, the expo and the dish station, each with its own productivity number. Round to whole shifts, then add a minimum crew for opening, closing and safety.

Two rules keep this honest:

Step three: turn demand into prep

Your POS already knows the item mix: what share of covers order each dish. Multiply the forecast by the mix to set prep quantities, then adjust for shelf life. If 22% of Friday dinner guests order the short rib and you forecast 286 covers, plan for about 63 portions, plus whatever buffer you can afford to waste.

Feed the same forecast into ordering, and purchasing stops being a guess as well.

Step four: measure forecast error every week

Compare forecast with actual for every hour and record the gap as a percentage. Average it over the week. Watch which hours and days miss most often, and look for the pattern behind them: weather, events, a new delivery channel or a menu change.

You do not need perfect accuracy. You need accuracy that improves, and a schedule that adjusts when it misses. Track these alongside it:

Pitfalls to avoid

When a spreadsheet stops being enough

A spreadsheet works for one location with one channel. It strains when you add locations, delivery and online ordering, frequent promotions, or when the forecast should flow into scheduling and purchasing without anyone copying numbers. At that point, connecting the POS, scheduling and ordering systems to one forecast saves hours each week and removes the copy-paste errors.

By the numbers

Published industry case studies from software firms show what better forecasting has delivered in hospitality. The figures below are self-reported by the firms that published them, and each describes a single project rather than an industry average.

Where to start

Pull the last six weeks of hourly sales for one location and build next Friday's forecast by hand using the method above. Then compare it with your current schedule. The gap is your first estimate of what forecasting is worth.

Delta Technologies connects POS, scheduling and purchasing data so forecasts drive the schedule and the order. See our services, read more on the blog, or book a systems review to map where labor and food cost are decided.