Controlling Food Cost with Recipe-Level Inventory

Most restaurants know their food cost percentage to the month. Very few know it to the dish. The difference matters, because an average hides the items that are quietly losing money on every sale.

Stock counting only answers half the question

A physical count tells you what is in the store today. Useful, but on its own it cannot tell you whether the amount you used was the amount you should have used.

For that you need two numbers. Actual usage is opening stock plus purchases minus closing stock. Theoretical usage is what your sales say you should have consumed, given the recipes. Where those two diverge, something is happening: over-portioning, waste, spoilage, staff meals not recorded, or theft.

Without recipes, you only ever have the first number, and no way to know whether it is reasonable.

Building recipes without boiling the ocean

Recipe-mapping the entire menu at once is how this project dies. Start where the money is.

Run an item-wise sales report for the last quarter and sort by revenue. Your top twenty items almost certainly account for most of your covers. Map those first. You will get the majority of the benefit from a fraction of the work, and the exercise gets easier as you build a library of ingredients and units.

Be honest about quantities. A recipe that says 150g of chicken when the kitchen actually plates 190g will show a variance that is really a data problem, and you will waste a week chasing a phantom.

Yields and prep are where it gets interesting

Raw purchase weight is not usable weight. You buy a whole fish and use perhaps half of it. You buy onions by the sack and lose skin, ends and the occasional bad one.

Your costing needs to account for yield, or every dish involving trimmed produce will look cheaper than it is. Where possible, cost from the usable quantity rather than the purchased quantity.

Batch prep deserves its own handling. If your kitchen makes twenty litres of gravy on Monday that feeds dishes all week, treating each portion as if it were made from raw ingredients at the point of sale gets messy. A batch recipe lets you produce the gravy once, book it into stock as its own item, and have individual dishes draw from it.

The reports that actually change behaviour

• Item-wise sales. Which dishes sell, in what volume. The starting point for everything else.
• Consumption against theoretical. The variance report. This is the one that finds problems.
• Purchase history by supplier. Ingredient prices move. A dish costed in January may be underwater by June if nobody re-checks.
• Wastage log. Only useful if staff actually record it, which means making it a ten-second job rather than a form.

A realistic cadence

Weekly counts on high-value and fast-moving items — proteins, dairy, alcohol. Monthly counts on everything else. Re-cost your top items whenever a major ingredient price moves, and at least quarterly.

The goal is not a perfect number. It is a number consistent enough that a change in it means something.

How MealNix handles it

Inventory items carry categories and units of measure, with stock tracked per branch and a full movement history. Recipes link menu items to their ingredients so that selling a dish draws the ingredients down automatically, and batch recipes cover prep made ahead. Purchase orders and supplier records keep the buying side in the same system, and the inventory and batch reports give you consumption and valuation without a spreadsheet.