Food cost and recipe costing: how to price a dish properly

What food cost percentage actually means

Food cost percentage is the share of a dish's selling price consumed by its ingredients:

Take a paneer dish selling at ₹320. If the paneer, butter, tomato, cashew and cream in it cost ₹103, your food cost is 32% and your gross margin is 68%. That ₹217 difference is what pays your rent, salaries, electricity, gas, aggregator commission — and only what's left after all of that is profit.

What's a healthy target?

For Indian full-service restaurants, a food cost between 25% and 35% is the usual working range. Where you should sit inside it depends on the category:

Notice the implication: a menu is a portfolio. Beverages and vegetarian staples subsidise the mutton and the prawns. That's fine and normal — but only if you know which items are doing the subsidising.

How to cost a dish, step by step

Costing one dish properly takes about ten minutes. Do it once for your top twenty sellers and you'll usually find at least two surprises.

1. 1 · Write down the actual recipe by weight. Not "some paneer" — 200 g paneer, 30 g butter, 100 g tomato, 20 g cashew, 30 ml cream. If two cooks plate it differently, cost the heavier version, because that's what you'll often serve.
2. 2 · Use your real purchase price, including GST. Most restaurants are on 5% GST with no input tax credit, so the tax you pay on supplies is a genuine cost. Cost from the invoice total, not the pre-tax figure.
3. 3 · Convert to a per-gram or per-ml rate. Paneer at ₹320/kg is ₹0.32/g. So 200 g costs ₹64.
4. 4 · Add yield loss. You buy 1 kg of onions but only 850 g reaches the pan after peeling. If you ignore trim waste your costs are understated by 10–20% on fresh produce.
5. 5 · Add the invisible items. Cooking oil, salt, spices, garnish, and the packaging on a delivery order. Individually trivial, collectively 3–8% of the plate.
6. 6 · Divide by the selling price. That's your food cost percentage.

Pricing backwards from a target margin

Once you know a dish's cost, don't guess the price — derive it. If you want a 70% margin (i.e. 30% food cost):

So a dish costing ₹103 should sell around ₹345 to hit a 70% margin. If you're currently charging ₹320, you're running about 68% — close, but you're leaving roughly ₹25 per plate on the table. On 15 plates a day that's over ₹11,000 a month from one menu item.

Round sensibly afterwards. ₹345 is fine; ₹343 looks like a spreadsheet escaped.

Why your bestseller is often your worst margin

Popularity and profitability are different things, and they're easy to confuse because both feel like success. A dish can be your top seller precisely because it's underpriced.

The useful exercise is to plot every item on two axes — how often it sells, and what it earns per plate. Four groups fall out:

• High volume, high margin — protect these. Never let quality slip, never discount them.
• High volume, low margin — your most urgent problem. A small price rise or portion correction here moves real money, because it multiplies across many covers.
• Low volume, high margin — promote them. Staff recommendations and menu placement can shift these cheaply.
• Low volume, low margin — candidates for removal. They occupy menu space, prep time and fridge space for very little return.

Costing is only true on the day you do it

This is the part that undoes most manual costing exercises. Onion prices triple in a bad season. Paneer, oil and tomato all move. A spreadsheet you built in January describes January.

There are two ways to keep it honest. Either re-cost your top items every quarter by hand, or map recipes to inventory once and let purchase prices flow through automatically — so when paneer goes up, every dish containing paneer re-costs itself and you can see which ones fell below target.

A quick reality check on theoretical vs actual

Recipe costing tells you what a dish should cost. Your actual consumption is usually higher, and the gap is where money leaks: over-portioning, wastage, spoilage, staff meals, and pilferage.

Compare the two. If your recipes say you should have used 8 kg of paneer this week and your stock count says 11 kg left the fridge, that 3 kg variance is worth investigating — and it's a far more useful number than a monthly P&L that only tells you something went wrong.