Your recipe costing says 29 percent. Your bookkeeping says 33.5. Both are right — and that gap holds more information than any other number in your business.

The month closes, the food cost ratio is higher than planned again, and you run through the usual explanations. Suppliers got more expensive. The accountant calculates differently. The recipe costing needs an overhaul.

All of that can be true. None of it explains why the gap stays roughly the same size, month after month.

Two numbers that measure different things

Target food cost comes from your recipe costing: dishes sold multiplied by their recipe value, added up. This number describes a world where every portion matches the recipe exactly, nothing spoils, and no one eats off the books.

Actual food cost comes from stock movement: opening inventory plus purchases minus closing inventory. That's what was actually consumed.

Run through our model operation — 60 seats, €100,000 monthly revenue: dishes sold amount to €29,000 by recipe. Stock movement shows €18,000 opening inventory, €34,200 in purchases, €18,700 closing inventory. Actual consumption: €33,500.

The gap is €4,500. In one month. For a business that might clear around €80,000 in annual profit, that's an order of magnitude that decides whether the whole thing is worth it.

Neither number is wrong. They just measure different things.

Where the money actually goes

In most operations, the gap splits across the same four causes, roughly in this order.

The biggest is portion drift, and it's also the least visible. Fried chicken with 180 grams of meat in the recipe gets plated by eye at 200 grams. Eleven percent more of the plate's most expensive component, across 400 portions a month. Nobody notices, because no single plate looks short.

Then staff meals. Cooked, eaten, never logged. Ten staff meals a day at three euros in food cost is €900 a month that shows up in consumption and never in revenue.

Spoilage and breakage rank third and are usually smaller than assumed, as long as order quantities match demand.

And finally, comps and remakes. A returned dish gets replaced. The replacement shows up in consumption but only counts once in revenue.

The gap isn't an error message — it's the metric

The point isn't which of these items is biggest for you. The point is knowing the gap exists at all.

Anyone who only sees actual food cost from the books knows, at month's end, that it was too high. Not where. So they reach for what's at hand: raise prices, or squeeze suppliers. Either can be correct. Both miss if the cause sits in the kitchen, not in the price.

Put target and actual side by side, and you have a number you can actually manage. It never hits zero — two to three percentage points is normal. But its movement means something. When it grows, something happened. When it shrinks, a fix worked.

Getting started

Take one closed month and calculate both figures once. The first time you see the number, you'll know more about your business than most of your competitors know about theirs.

Don't run it for three hundred items — run it for the twenty that make up most of your food cost. That's where the gap lives anyway.

And repeat it monthly. A number you see once a year is statistics. One you see monthly is control.

Someone has to watch when no one has the time

Now the uncomfortable part: running this calculation every month takes time that doesn't exist between scheduling, complaints, and hiring. That's why so few do it.

ORA, our AI agent at optimizes Intelligence, calculates the gap continuously and flags it the moment it moves. Not as a number in a dashboard someone would have to check, but as a concrete recommendation:

Recommendation from ORA

Actual food cost in May stands at 33.5%, against a target of 29.0%. The gap has widened by 0.8 percentage points versus April.

Notable: for fried chicken, calculated consumption runs 14% above recipe, for beef soup 11%. Both items sold above average in May.

Suggestion: portion-control these two dishes for two weeks using a scale, then re-measure.

That's the difference between a report that exists and one that changes something.

Let's work through your gap together. Book thirty minutes, we'll apply this article's framework to your own numbers, and you'll know afterward exactly where your difference sits. Schedule a call.

Next up: the special case we skipped here — what happens when it's not the actual value that's too high, but the target value that's too low.