How to Calculate Food Cost Percentage and Price Menu Items

How do you calculate food cost percentage, step by step?

Food cost percentage tells you what share of a dish's price is spent on its ingredients. The formula is simple, but the inputs need care, because most costing errors come from units, yields and forgotten items rather than arithmetic. Follow these steps for one dish, then repeat for the menu.

• 1. Write the standard recipe: every ingredient and its weighed quantity for one portion, as the cook actually makes it.
• 2. Convert each pack price to a cost per recipe unit: pack price divided by pack size in grams, millilitres or pieces.
• 3. Divide by the yield %: the usable share after trimming, peeling, boning, shelling or thawing, from a yield test.
• 4. For batch items (gravies, marinades, doughs, rice), divide the batch cost by the finished batch weight.
• 5. Multiply each usable cost by the recipe quantity and add the lines.
• 6. Add a garnish and sides allowance to get the plate cost; add packaging for takeaway and delivery.
• 7. Divide the plate cost by the menu price without tax. That is the food cost %. Subtract the plate cost from that price for the gross profit per dish.

What does a worked example look like?

Here is one dish costed from a fictional composite Indian-fusion restaurant, in Indian rupees with prices shown without tax. Chicken thigh costs 330 a kilo as purchased with a 95% yield after trimming, so a usable gram costs 0.347. The makhani gravy is a batch: about 795 rupees of ingredients that finish at 4.5 kg, so 0.177 a gram.

Read the result in two ways. The food cost is 29.8%, just under a 30% target, so the percentage looks healthy. The gross profit per dish is about 309 rupees, which is above the average for the curries in this menu, so the dish also earns its place in cash. If chicken thigh rose 10%, the plate cost would rise by about 7 rupees and the food cost to about 31.4%: a small change per plate that adds up quickly on the restaurant's best seller.

Why does yield matter so much?

Yield is the usable share of an ingredient after preparation. Shell-on prawns that lose 45% of their weight when peeled and deveined have a 55% yield. If they cost 680 a kilo as purchased, a usable gram costs about 1.24, not 0.68. In the same composite restaurant, a tandoori prawn dish costed without the yield looked like a 30% food cost; with the yield it was nearly 53%.

Run each yield test three times on different deliveries and use the average. Yields change with supplier, grade, season and the person doing the prep. Cooking loss works the same way for anything that shrinks or grows in the pan: weigh the finished batch and cost per finished gram. Rice works the other way: one kilo of raw basmati may give about two and a half kilos cooked, so a cooked gram costs far less than a raw one.

Is a low food cost percentage always better?

No. Food cost % shows the share; gross profit per dish shows the cash. In the composite example, dal makhani has a 13% food cost and keeps 252 rupees a plate, while tandoori prawns have a 53% food cost and keep 292. Chasing the lowest percentage would push guests towards cheaper dishes and leave less cash to pay rent and wages.

Set a target food cost % for the whole menu from your own first eight weeks of data rather than a published range, and read each dish's percentage beside its cash margin and its popularity. Prime cost, food and drink cost plus labour, is the wider view: a cheap dish that takes a skilled cook twenty minutes is not cheap.

How do you set a menu price from the food cost?

Use four methods together. Cost-plus divides the plate cost by your target food cost %. A cash margin target adds the gross profit the dish should earn, often near its menu section's average. Market pricing compares similar dishes nearby and on the apps, using prices and portions you collected yourself. And the dish's role (signature, add-on, dietary option) can justify an exception.

Be careful with menu psychology. Some findings are well known but rest on narrow evidence. A Cornell study by Sybil Yang, Sheryl Kimes and Mauro Sessarego found that 201 diners at a single restaurant spent more when prices were shown without a dollar sign; that is a reason to test, not a rule. Keep the price format consistent and follow local rules on showing tax and fees.

Combos and meal deals deserve their own check. Add the plate costs of the items, decide the cash the combo must keep, set the price, and only then work out the saving to the guest and state it honestly. A dessert or side with a low plate cost is often the cheapest way to make a combo feel generous. Round prices to a consistent step that suits your currency, such as the nearest 10 rupees, 0.50 pounds or 0.25 dollars.

How do you price dishes for delivery apps?

Work out what each dish keeps on each app: the app price without tax, minus commission, payment fees and the discounts you fund, minus plate cost and packaging. Compare it with the cash the dish keeps at its dine-in price. In the composite restaurant, using illustrative deductions of 33%, butter chicken kept about 301 rupees in the dining room at 440 and about 172 on the app at 480.

Then decide a floor, for example keeping at least 70% of the dine-in cash margin on your main app sellers, and check whether a delivery-specific portion or container makes more sense than a higher price. Commission rates, discount funding and any rules about price differences between channels are set by your contract and the platform's current terms, so check them directly. Test every discount campaign by calculating how many extra orders it needs just to match a normal week. In the composite example, a 40% offer capped at 120 rupees and fully funded by the restaurant cut the cash per order from about 290 to about 202, so it needed roughly 43% more orders just to stand still.

How does menu engineering use food cost?

Menu engineering, developed by Michael Kasavana and Donald Smith, judges each dish on popularity and contribution margin (price minus plate cost) together. A dish is popular if its share of its section's sales reaches 70% of a fair share; its margin is high if it beats the section's weighted average. Stars are high on both, plowhorses popular with lower margins, puzzles profitable but slow, dogs low on both.

Each class suggests a first move: protect stars, trim plowhorses' costs before nudging their prices, make puzzles visible through placement and description, and replace or re-specify dogs unless they have a clear job on the menu. The classes start the conversation; kitchen capacity, labour time and shared ingredients finish it. Use at least four weeks of sales, add units from the POS and every delivery app, and leave out comps and staff meals so the popularity figures reflect what guests actually chose.

What are the most common food costing mistakes?

These are the errors that make a menu look healthier on paper than it is in the bank. Most of them are invisible on a single plate and obvious across a month of sales, which is why the weekly comparison of theoretical and actual food cost matters as much as the costing itself.

• Costing from a supplier's list price instead of the latest invoice.
• Ignoring yield on trimmed, peeled, boned or shelled ingredients.
• Forgetting cooking loss on gravies, rice and roasts.
• Leaving out oil, garnish, sauces, sides and delivery packaging.
• Judging dishes on food cost % alone.
• Adding one flat uplift to every app price.
• Costing once at opening and never again.
• Comparing your percentage with someone else's instead of with your own theoretical cost.

How do you check the food cost is real, week by week?

Compare theoretical food cost (units sold multiplied by each dish's plate cost) with actual food cost (opening stock plus purchases, minus credits and closing stock). The gap points to receiving errors, over-portioning, recipe drift, waste, counting errors, stale costings or unrecorded comps. Waste alone is significant: UNEP's Food Waste Index Report 2024 estimated 1.05 billion tonnes of food wasted in 2022, 28% of it in food service, and WRAP's 2013 study of UK hospitality found 45% of outlet food waste arose in preparation.

Set a tolerance, investigate anything above it in a fixed order, and re-cost every dish when an ingredient price moves past a trigger such as 5%. A twenty-minute weekly review of four numbers (actual food cost %, the gap to theoretical, delivery share of sales and dishes re-costed) keeps the habit alive. When a supplier raises a price, a re-cost is not automatically a price rise: ask for a fixed price period or a volume discount, get a second quote for the same specification, consider a different grade or pack size, adjust a garnish or portion and update the recipe card, and reprice only after checking every channel. The Menu Costing and Engineering System from Bodhih Training includes a workbook that runs these calculations, an aggregator price calculator and the review forms. If you want to check the arithmetic and data skills behind this work, the Business Arithmetic assessment on AssessAll is one way to measure a starting point, and Jobulary shows how to turn gaps into a development plan.