Pull up your menu. Look at your three best sellers and tell me, to the dime, what each one costs you to make this week.
Almost nobody can. I own a restaurant and two trucks, and for years I could not either. We priced the menu when we opened, we nudged a few things when something got obviously expensive, and the rest of the time we ran on a number from a spreadsheet somebody built in a different economy.
That gap is where the money goes. Not in one dramatic hit, in a few cents a plate, a few hundred plates a week, every week, for a year.
Your cost moves weekly. Your price does not.
Beef moves. Oil moves. Cheese moves hard and nobody tells you. Your invoices change most weeks and your menu changes once a year, if that, usually when you finally panic and reprint everything at once.
Every week between those two events, you are selling at a price you set against a cost that no longer exists. If the cost went down you never captured it. If it went up, and it mostly went up, you ate it.
The restaurants getting ahead right now are not smarter than you. They just shortened that gap from a year to a week, and the only reason they can is that the counting finally got cheap.
The pricing audit, and what it usually finds
This is the one to run first, and you can run it this week. Take your ten best sellers. Pull your current invoice prices. Cost out every plate, properly, including the things people skip: the sauce, the garnish, the oil, the paper.
Two things tend to fall out of it, and the second one hurts.
First, two or three items are underpriced badly enough to be visible once you see the number. Usually something you have not touched since you opened because you are attached to it.
Second, the dish you are proudest of is often your worst margin, and it is frequently the thing you push hardest. The special you talk up, the one the servers mention first, the one on the sign out front. You have been spending your best sales effort on the plate that pays you least.
That is not a reason to kill it. It is a reason to know it, so you can price it, shrink the portion slightly, or deliberately use it as the loss leader instead of accidentally.
Inventory that counts itself
Same story one step upstream. Most places set pars once and then run on memory and vibes, which means you are either sitting on product you do not need or you are 86ing something on a Friday.
Counts and order sheets that adjust to what you actually sold, by week and by season, fix both ends. Less cash sitting in the walk-in, fewer missed sales, and less of what nobody counts, which is waste.
Waste is the one that stays invisible. It does not appear on a report. It goes in a bin at eleven at night and nobody writes it down. The first time you count it honestly for two weeks, it is a bad evening.
Why this is urgent and not just a good idea
Because it compounds, in both directions.
Shave a few points of food cost and the money shows up every week from then on. It is the only lever in this business that keeps paying without you selling a single extra cover.
Leave it and the reverse happens. You absorb increases quietly, your margin thins, and you do not notice until a slow month arrives and there is no cushion under it. Then you are repricing in a panic, raising everything at once, and your regulars feel it in one jump instead of not noticing it over four small moves.
And somebody in your market is already doing this monthly. Not because they bought something clever. Because they sat down one Sunday and started.
The hour that starts it
Do not buy a platform. Take your last three invoices and your top ten sellers, sit down with an AI tool you already have access to, and cost the plates out. It will do the arithmetic and catch the things you forget, like the cup and the lid.
Then do the same with your sales mix: what sold, what died, what the weather did to it. You are looking for the gap between what you push and what pays.
Second time around takes twenty minutes. By the fourth time your manager can run it, which is the actual goal, because a job only you can do is not a system, it is another shift.
What to do Sunday
Ten items. Current invoices. Honest costing including the small stuff. Then look at your menu and decide which three prices move, and whether the thing you push hardest deserves the push.
That is the whole exercise, and it is worth more than any marketing we could sell you until it is done.
Next: Just because you can build it does not mean you should