Restaurant reports worth your time: six numbers to review every week

The rule: a report that changes no decision is not worth reading. For a restaurant or café it is enough to review six numbers once a week: the mix of order types, waiter sales, discounts, cancellations, kitchen time, and the most-ordered options. Fifteen minutes, in the same order every time.

1. Order types: where does your money come from?

How much of your revenue is dine-in, and how much takeaway, delivery and counter? This number decides where you invest. If delivery is a third of your revenue and you treat it as a sideline, you are neglecting a third of your restaurant. If it is 5% and eats a whole cook at peak, that is a different question.

The decision: how you allocate staff, space and marketing.

2. Waiters: who sells, and who only serves?

Bills, guests, total and tips per waiter. Do not compare totals alone — whoever works Thursday evening will win. Compare average spend per guest: the waiter’s total ÷ their guests. A gap here means one suggests starters and dessert while another just writes things down.

The decision: who trains whom. Your best waiter is your training programme.

3. Discounts: how much do you give, and who gives it?

A discount is a legitimate marketing cost when it is a decision, and a leak when it is a habit. The report shows who gave discounts, on how many bills and for how much. One employee giving half the restaurant’s discounts is not necessarily doing wrong — but deserves a question.

The decision: the discount ceiling per role, and when manager approval is required.

4. Cancellations: the number everyone avoids

What matters most is telling a cancellation before the item went to the kitchen (the guest changed their mind — no cost) from one after (food prepared and not paid for). The second kind is either an order-taking mistake, a quality problem, or something worse. The reason recorded with each cancellation is what tells them apart.

The decision: if “wrong order” keeps recurring, the fix is in how orders are taken. If “too slow” recurs, look at the next number.

5. Kitchen performance: where is the bottleneck?

Average time from an item being sent to being ready, per station. An overall average is useless; what helps is knowing the grill takes 18 minutes and the bar 4. A slow station is not necessarily lazy — it may be loaded with too many items, or a person short at a particular hour.

The decision: re-routing items between stations, or an extra person at peak. This number does not exist without a kitchen display recording when things are ready.

6. Options: what do guests actually add?

Which extras get ordered, and how much do they bring in? If “extra cheese” goes with a third of your burgers, it may be time for a cheeseburger at a higher price. If an option has not been ordered in two months, it slows the waiter for nothing.

The decision: pricing extras, and tidying the menu.

The fifteen-minute routine

Pick a fixed day — Monday morning, say — and do not go past fifteen minutes. The aim is to notice change, not to study the figures.

In Sahl

Sahl’s restaurant module has nine reports: order types, waiters, tables, stations, options, tips, discounts, cancellations (with their reason and whether they came after sending), and kitchen performance. They open from the control panel and the app, for any period. On the accounting plan you also get financial reports from the same books. For everyday figures in general, read daily sales numbers.