Most restaurants check sales. Fewer check the four numbers sitting right next to it that actually explain why sales look the way they do. On their own, each number is a fragment. Together, they tell you whether the week actually went well, or just looked like it did on the surface. Here are the five worth a proper look every week, what each one is really telling you, and what a bad reading actually looks like in practice.
1. Sales, compared, not just totalled
What it is: This week’s sales against last week’s, and against the same week last year.
Why it matters: A number on its own tells you nothing. £8,000 this week is either a great result or a worrying one, entirely depending on what it’s being compared to. Trend matters more than total.
What it looks like in practice: A steady Thursday that’s actually 12% down on the same Thursday last year — invisible if you’re only looking at this week in isolation, obvious the moment you put the two side by side.
Watch for: A quiet week that looks fine on its own but is actually down against last year — the kind of drift that’s easy to miss until it’s a real problem, because no single week ever looks alarming by itself.
2. Profit by category, not just revenue
What it is: Which menu categories are actually making money, not just moving volume.
Why it matters: Revenue and profit are very often different questions. A category selling well can still be quietly dragging on margin if food cost has crept up without anyone noticing — a supplier price rise that never made it back into the menu price is one of the most common, least visible profit leaks in hospitality.
What it looks like in practice: Your best-selling main is still your best-selling main, but its margin has slid three points over the last two months because a key ingredient’s gone up in price and nobody adjusted for it.
Watch for: A best-seller with a margin that’s thinner than you’d assume — worth checking against actual cost data, not gut feel or last year’s numbers.
3. Voided and cancelled orders
What it is: How many orders were voided or cancelled this week, and by whom.
Why it matters: A small, steady number is normal — genuine mistakes happen, every shift, everywhere. A sudden spike, or a pattern tied to one till or one shift, is usually the first sign of either a training gap or something worth a closer look.
What it looks like in practice: One terminal showing three times the void rate of every other till on the same shifts — sometimes it’s a training issue, sometimes it’s worth a direct, calm conversation.
Watch for: Cancellations clustering around a specific time, terminal, or member of staff, rather than being spread evenly across the week.
4. Stock variance
What it is: The gap between what your system says you should have in stock, and what you actually have.
Why it matters: Some variance is normal — spillage, portioning drift, minor waste, all part of running a kitchen. A growing gap, week over week, usually means something specific: Over-portioning that’s crept in gradually, waste that isn’t being logged properly, or stock going missing somewhere between delivery and plate.
What it looks like in practice: A 2% variance on a high-cost ingredient might be nothing. The same 2% appearing three weeks running, on the same ingredient, is a pattern worth actually investigating rather than writing off as normal wastage.
Watch for: The same ingredient showing variance two or three weeks running — a one-off is noise, a repeat is a signal.
5. Repeat customer rate
What it is: What share of this week’s customers have been in before.
Why it matters: New customers are visible and satisfying to chase — every marketing push is built around them. Repeat customers are usually the more profitable, more reliable number, and the one that quietly tells you whether the actual experience is bringing people back, rather than just getting them through the door once.
What it looks like in practice: A busy month on paper that’s almost entirely new faces, with barely any of last month’s customers coming back — a growth number that’s actually masking a retention problem.
Watch for: A healthy footfall number masking a low repeat rate — busy isn’t the same as building a base, and the two can look identical on a simple sales report.
How to actually make this a five-minute habit
The reason most operators don’t check these numbers regularly isn’t that they don’t care — it’s that “check the numbers” is vague enough to keep getting pushed to tomorrow. A few things that make it actually stick:
- Pick a fixed day and time, ideally the quietest admin moment in your week, and treat it the same as any other recurring task
- Look at all five together, not just whichever one happens to catch your eye — the value is in the combination, not any single number in isolation
- Write down what surprised you, even briefly — a number that looks the same as last week isn’t worth dwelling on, but anything unexpected is worth a note so you can check whether it’s a blip or a trend
Where these numbers actually come from
None of this requires a spreadsheet built from scratch. YUMA’s back office already generates sales, profit and loss, inventory, and customer reports as standard, with a dashboard view pulling the headline numbers together so a weekly check takes minutes, not an afternoon reconciling separate systems.
The takeaway
Sales alone tells you what happened. These five numbers, together, start telling you why — and where to look next. None of them need a data analyst to check. They need five minutes, once a week, a fixed slot in the diary, and the discipline to actually look before the numbers surprise you rather than after.
Want to see what a weekly check actually looks like on a real dashboard? Book a demo with YUMA and we’ll show you.
