Restaurant Food Cost: How to Calculate It and Lower It Without Losing Quality

What food cost is and why it matters

Food cost is the share of food revenue that goes on the ingredients used to produce the dishes you sold. It reacts faster than almost any other cost line: one wrong supplier price or one sloppy cook can move it within a week.

The number is only useful if it is calculated the same way every period. The goal is not the lowest possible figure but a stable, predictable one that matches your concept and pricing, so you notice immediately when something drifts.

The formula: actual food cost from inventory

The basic formula is: food cost % = cost of ingredients used / food sales × 100. The cost of ingredients used is not the same as what you bought. It is calculated as opening inventory + purchases for the period − closing inventory. Stock sitting on the shelves at month end is not a cost yet; stock that was on the shelves at the start and is now gone is.

Keep food and beverages separate. Bar cost follows the same logic but has different benchmarks, and mixing them hides problems in both. Record staff meals and write-offs as separate lines.

Define the scope before calculating: one kitchen or several locations, which sales count as food revenue, and when the period closes. Deliveries, inventory and sales must belong to the same period. Apply one consistent rule to customer discounts and refunds; total restaurant revenue is not the right denominator for a kitchen-only calculation.

When ingredients move between locations, the basic expression needs adjustment. For a management reconciliation of one kitchen, use: usage = opening inventory + supplier receipts − returns to suppliers + transfers in − transfers out − closing inventory. Value all entries using an agreed method. If the system already reports receipts net of supplier returns, do not deduct those returns again.

A waste entry explains usage; it is not stock still on the shelf. When actual usage is already derived from physical counts, do not add the same recorded waste a second time. Show separately how much usage is explained by waste, staff meals and other recorded operations. If you exclude these items from a kitchen metric, name that metric clearly and retain a reconciliation to total usage.

How to take inventory properly

Inventory is the foundation: if the closing count is wrong, the food cost is wrong. Count at the same time on the same day of each period, ideally before a delivery and after service. Use a count sheet that follows the storage layout, shelf by shelf.

Count in consistent units and value inventory using one documented method agreed with the person responsible for your accounting system. Apply the same method to opening and closing inventory; changing the valuation method can change the reported food cost without changing kitchen performance. Include semi-finished products such as sauces, stocks and marinated proteins, valued by their recipe card. Counting in pairs, with a manager checking meat, fish and cheese, reduces errors.

Theoretical vs actual food cost, with an illustrative example

Theoretical food cost is what the ingredients should have cost based on the items sold and their recipe cards, assuming perfect portions and no waste. Actual food cost comes from inventory. The gap between them is the real management signal: waste, over-portioning, unrecorded write-offs, pricing errors or theft.

Illustrative example, not real data: opening inventory 4,000, purchases 11,000, closing inventory 3,500. Ingredients used = 11,500. Food sales = 35,000, so actual food cost = 32.9%. Recipe cards multiplied by items sold give 10,500, a theoretical food cost of 30.0%. The 1,000 difference, about 2.9 percentage points, is money to find, dish by dish.

Do not label the entire gap as a proven loss. Reconcile it to recorded waste, staff meals and other known uses, data errors, and the remaining unexplained variance. Record the evidence for each line. The same amount must not appear both as explained waste and as unexplained variance. Use that reconciliation to decide whether to recount stock, update a recipe card or check portions.

Recipe cards, yield and waste

A recipe card lists every ingredient of a dish with its quantity, unit cost and total plate cost. Without them there is no theoretical food cost.

The most common mistake is costing ingredients at purchase weight. A whole fish, a leg of lamb or a crate of herbs loses weight in cleaning, trimming and cooking. Run yield tests: weigh the product before and after processing, and calculate the usable yield. If a kilogram of salmon gives 650 grams of fillet, the real cost per kilogram of fillet is the purchase price divided by 0.65. Update cards whenever supplier prices change.

Portion control and kitchen discipline

Portion control is where theoretical and actual cost usually diverge. Scales on every station, portioning spoons and ladles of fixed volume, pre-portioned proteins and photos of the reference plate at each station remove guesswork.

Add simple routines: 1) a waste log where every write-off is recorded with a reason; 2) first in, first out storage with labels showing the date of preparation; 3) prep lists based on forecast covers rather than habit; 4) a weekly review of the dishes with the largest gap. Consistent portions protect both margin and the guest experience.

Supplier prices and purchasing

Ingredient prices move constantly, and categories move differently. The USDA Economic Research Service, for example, publishes separate forecasts for beef, eggs, produce and other categories, and they can diverge sharply within one year. So food cost can drift without any change in the kitchen.

Compare invoices with agreed prices on every delivery, weigh and check quality at receiving, and keep at least two suppliers for key items. Order to par levels instead of by feel. When a key ingredient rises sharply, decide consciously: renegotiate, substitute an equivalent product, adjust the portion or change the price.

Menu engineering basics and a sensible target

Food cost percentage alone can mislead. A steak at 40% may earn more money per plate than a salad at 20%. Menu engineering looks at two things for each dish: how many are sold and how much gross profit each one brings. Popular, high-profit dishes deserve the best menu placement; unpopular, low-profit ones are candidates for rework or removal.

As for targets, a range of roughly 30–35% is often cited as a rule of thumb, and Toast refers to National Restaurant Association data showing median food and beverage costs of around 32% of sales. Your own target depends on concept, format and price level. Ready-made accounting spreadsheets are available in the site's shop section.

FAQ: common questions about food cost

How do I calculate food cost for one dish? Divide the ingredient cost per portion from its recipe card by the dish’s selling price on the same tax basis, then multiply by 100. Illustrative figures: a portion costing 6 and selling for 20 has a food cost of 30%. Use one currency throughout. This measures ingredient cost, not profit after labour, rent and other expenses.

How often should I calculate food cost? At least monthly with a full inventory, and weekly for high-value categories such as meat, fish and dairy.

Can I lower food cost without cutting quality? Yes. Most savings come from reducing waste, fixing portions, recalculating yields and buying better, not from cheaper ingredients.

What if actual food cost is much higher than theoretical? Check the data first: missing invoices, wrong prices in recipe cards, counting errors. Then compare dish by dish, review the waste log and watch portioning during service.

Continue reading