Covers, Stock and Tips: System Data for Hospitality

Hospitality is the industry most often praised for running on instinct, and the one where instinct is most expensive. A chef who can taste a sauce and tell you what is missing can rarely tell you what the plate costs — not because they do not care, but because the answer is spread across a supplier invoice in a folder, a recipe in someone's head, a stock sheet from Sunday and a wastage note that never got written down. This is the seventh piece in our system data series, and hospitality may be the clearest case of all: the numbers exist, they are simply never in the same place at the same time.

A restaurant does not fail because one dish is priced wrong. It fails because nobody could see that it was priced wrong for eleven months.

What the paper actually looks like

Walk into most independent venues and the operating system is physical. A booking diary at the host stand, in pencil, which is also the only record of last December. Walk-ins that are never counted, so the busiest hours of the week leave no trace. Function enquiries in a WhatsApp thread, quoted from memory, half of them never followed up. A stock sheet counted on a Sunday night by whoever drew the short straw. Wastage on a clipboard by the bin, filled in honestly for about a week after every stock take. A roster on a whiteboard, rubbed out and rewritten. And tips, split by a formula that lives in the manager's head and is explained anew to every new starter.

None of this is negligence. Each piece was the fastest thing that worked at the time. The problem is that none of it can be added up, so the business gets managed on the bank balance, which is the slowest possible feedback loop.

Where the money actually leaks

Four leaks, and they are almost always the same four.

The gap between theoretical and actual. Your till knows how many G&Ts you sold. Multiply by the measure and you know how much gin should have gone. Compare with what actually went, and the difference is over-pouring, breakage, staff drinks, comps or theft — and until you can see the number weekly, you cannot tell which. That single comparison is the most valuable report in the business, and it is impossible without recipes and stock in a system.

Costs that moved while the menu stood still. The dish was costed when the menu was designed. Since then oil, dairy, protein and packaging have all moved, several times. If nobody re-costs when supplier prices change, the margin erodes silently — and the erosion is worst on your most popular dish, because volume multiplies the error.

Voids, comps and discounts. Individually defensible, collectively a line item nobody has ever seen totalled. A system that requires a reason code turns a vague generosity into a number you can manage.

Staffing that does not match the room. Rostering by habit means over-staffed Tuesdays and a Friday where the kitchen drowns and the reviews say "slow service". Covers by hour, by day of week, is the data that fixes it, and it is free the moment the till is capturing properly.

The till is the capture point

Every hospitality business already has the perfect data-capture device: the point of sale, touched hundreds of times a day by people who have to touch it anyway. That is the principle behind capturing data at the source — the record gets made where the work happens, not written up afterwards by someone with a clipboard. A till that is set up properly gives you covers, spend per head, item mix, time of sale and server, without asking anybody to do a single extra thing. A till set up badly — half the menu on a "sundries" button, modifiers as free text — gives you a number that reconciles to the cash and tells you nothing else. The difference is a week of setup, once. We have watched this transition up close in a retailer's move from a manual till to a real POS, and hospitality gains even more from it, because the recipe layer sits underneath.

Recipes are the missing link

The piece that turns a till into a management system is recipe costing: every menu item broken into its components with quantities, each component linked to what you actually pay for it. Do that, and three things become automatic. Every sale depletes stock at ingredient level, so your stock count becomes a check rather than a discovery. Every supplier price change re-costs every dish that uses it, and tells you which ones just fell below target margin. And every dish carries a real gross profit number, so the menu can be managed instead of guessed.

You do not have to do the whole menu. Start with the twenty items that make up most of your sales, and the top ten bar lines. That is a weekend of work and it covers the large majority of your food and beverage spend.

You cannot manage a menu you have not costed. Until each dish has a number, "our margins are tight" is a feeling, not a finding.

What you can finally see

Once covers, recipes and stock are in one place, the questions that used to be unanswerable become a report you glance at on a Monday. Which dishes are both popular and profitable — and which are popular and barely break even, which is the one every kitchen has and nobody has proved. What the variance was on spirits this week. Covers per hour by day, against hours rostered. Wastage by category, which is usually the fastest saving available to any kitchen. Average spend per head, and whether that upsell push actually moved it. No-show rate, and what a deposit policy on Friday and Saturday bookings would be worth.

Bookings, no-shows and the customer you cannot see

A digital booking system is not about looking modern. It is about knowing your no-show rate, being able to take a deposit on peak sittings, keeping the history of a regular's table and allergy, and — quietly the most valuable part — knowing how many covers you turned away because you were full. That last number never appears anywhere in a paper diary, and it is the one that tells you whether to add a sitting, extend the terrace or hold the line on price.

It also comes with an obligation. The moment you keep names, numbers and dietary information, you are processing personal information under POPIA, which means marketing to that list requires consent and a way out. Worth setting up correctly on day one rather than unpicking it later.

The bit that decides whether any of this works

Hospitality has high staff turnover, split shifts, and a service period where nobody has a spare hand. A system that requires a calm moment will not be used, because there is no calm moment between six and nine. So: capture happens on the device that is already in their hand, wastage takes one tap and a reason, and the manager checks a short report at the end of service rather than filling in a spreadsheet at midnight. If it takes longer than the paper it replaced, it will lose — which is the whole argument of adoption is the project, and nowhere is it truer than in a kitchen mid-service.

Get it right and the change is not that you have more data. It is that on the first Monday of the month you can say, with evidence, that the special everybody loves has been losing money since March, that Tuesday is over-staffed by two, and that the gin variance stopped the week you started counting. None of that is instinct. All of it was already happening — it just had nowhere to be written down.