Standard recipes and portion cost: how hotel kitchens price a plate

A standard recipe card fixes the ingredients, usable quantities, portion size and yield for a stated number of portions. A portion cost is each ingredient's usable quantity multiplied by its cost per usable unit, added up and divided by the portions the recipe yields. Yield tests convert purchase weight into usable weight.

Two kitchens can buy the same beef at the same price and sell the same steak at the same menu price, and one of them loses money on it. The difference is rarely the chef. It is whether anyone wrote down what the plate actually contains. After reading this, you will be able to write a standard recipe card, run a yield test, build a portion cost line by line, and keep both current when supplier prices move.

What does a standard recipe card contain?#

A standard recipe card is the fixed, written version of a dish: exact ingredients, usable quantities for a stated number of portions, the method, the portion size and the cost of one portion. It is the unit the kitchen, the storeroom and the cost controller all count in.

Standard means agreed once, used by everyone and changed only deliberately. It is not the same as a chef's personal recipe, which lives in memory and drifts with each service. The card is what allows the standard recipe cost hotel teams rely on to mean the same thing in every department.

Two details decide whether the card works. First, the quantity written on the card is the usable weight that reaches the plate, not the weight bought from the supplier; the purchase weight belongs on the yield line, not in the ingredient line. Second, sub-recipes such as sauces, stocks, doughs and marinades get their own cards and are referenced by name, so a sauce used in six dishes is costed once and a change to it updates all six.

Version control is the part most kitchens skip. Without a version number, a date and an approver, two copies circulate and nobody can say which one the cost was built on. Give each card one owner: the chef writes it, the cost controller costs it, the food and beverage manager approves it. Shared ownership is how cards stop being updated.

The card is one step in a wider cycle of purchasing, receiving, issuing and stocktaking, which the guide to food and beverage cost control in hotels covers end to end.

Why does yield matter, and how do you run a yield test?#

Because you buy a whole item and cook a part of it. Yield is the share of the purchase weight that reaches the plate, and it is the number that turns a supplier price into a usable cost.

Three terms, in plain words. As-purchased weight is what the invoice and the delivery note describe. Edible portion weight is what the cook can use. Yield percentage is edible portion divided by as-purchased weight.

Run the test the way the kitchen actually works: weigh the item as received, break it down with the same trim the section uses, weigh the usable parts, and weigh the trim separately. Trim is not automatically a loss. If bones go to stock or trim goes to staff meal, record that use and credit it; if it goes in the bin, it is a full loss.

Cook loss is a separate number from trim loss. A roast shrinks in the oven, so if the card is written on cooked weight, the test must be run on the cooked item. Repeat the test three times and take the average; size, season and supplier change the result, and a single test is an anecdote.

Test what matters: meat, fish, fresh produce and any item whose price or trim varies. Stable dry goods can use supplier or reference figures, provided the card says that is what it did. And yield starts at the door: if the receiving procedure does not weigh and record what arrived, the test is measuring a delivery nobody verified. The guide to hotel purchasing and receiving, from order to storeroom covers that step.

For example, here is an illustrative yield test with invented figures:

The cost per usable kilo is the purchase price divided by the yield percentage.

How do you build a portion cost, line by line?#

Multiply each ingredient's usable quantity by its cost per usable unit, add the lines, then divide by the number of portions the recipe yields. Add nothing you cannot name on the card.

Cost per usable unit is the purchase price divided by the yield percentage. For example, a 400 lira kilo of tenderloin at 65 percent yield is about 615 lira per usable kilo, and that is the number the card uses. The price must come from the latest invoice, not from memory or last season's list; an old price produces a confident and wrong portion cost.

Sub-recipes enter as a line: the cost of the sauce batch divided by the share of the batch the dish uses. Decide once whether labour, energy and overhead sit in the portion cost or outside it, and write the choice on the card. Two dishes costed on different models cannot be compared.

The common errors are predictable: costing purchase weight instead of usable weight, costing a recipe written for ten portions and dividing by twelve, forgetting the bread, the amuse or the oil in the fryer. Keep decimals inside the lines and round only the final portion cost, and state the rounding rule so two people costing the same dish get the same answer.

Cross-check monthly: portion cost times portions sold should roughly match the value of what was issued from the storeroom. A large gap means the card or the issuing is wrong, and both are worth finding.

For example, a worked cost for a dish yielding 10 portions, with invented prices:

Total: 972.00 lira for 10 portions, so the portion cost is 97.20 lira. The 1,800 g of tenderloin is usable weight; at 65 percent yield the kitchen needs roughly 2,770 g as purchased.

How do you keep portion costs current when supplier prices change?#

Keep every ingredient price in one list and let the cards point at it, so a price change updates every dish that uses the ingredient instead of only the ones someone remembered to fix.

One price list, one owner: each ingredient with its unit, its supplier and the date of the invoice the price came from. Cards reference the ingredient, never a number typed in by hand. Set review triggers rather than a vague intention: a new invoice price that moves past a threshold the hotel chooses, a monthly pass over the highest-spend items, and the change of season.

Re-costing is not re-pricing. Updating a cost produces a list of dishes whose portion cost moved; changing a menu price is a separate decision with the guest and the market in it. Keep dated versions either way. If a guest or an auditor asks in April what a dish cost in March, the answer should be a card, not a recollection.

A spreadsheet is enough if one person owns it and it is versioned; the usual failure is three copies on three laptops, each slightly different. Keep the tax basis consistent across all cards, whether prices are held with or without value added tax, and confirm the treatment with your accountant rather than assuming it.

How do recipe cards work for buffets and batch production?#

The same card, scaled. A buffet item is costed per batch with a stated number of portions, and the buffet's cost is the sum of the batches actually produced, not the sum of what guests ate.

Write the multiplier on the production sheet so the kitchen and the cost controller work from the same number; a verbal multiplier is a different recipe every service. Plan from forecast covers per item, then set the batch size, then record what was produced and what was left at the end of each service.

Leftovers have three destinations: reused within the hotel's own food safety rules, used for staff meal, or discarded. Each must be recorded, because unrecorded leftovers are exactly the gap between the card and the storeroom. Cost per portion produced and cost per cover served are different measures; when the buffet is inside the room rate, the second one is the useful figure, and it belongs to a revenue discussion rather than this one.

Buffet cards drift faster than à la carte cards because quantities are estimated by eye. Write gram-based batch cards and weigh the batch, not the plate. Watch the expensive items on the line, meat, seafood and cheese, on their own cards, since a small drift there outweighs a large drift on bread.

How should recipe costs feed menu decisions?#

As evidence, not as an automatic price. The card tells you what a dish costs, the sales record tells you how often it sells, and a person decides what to do with both.

Put two numbers side by side per dish: portion cost and units sold. That simple pairing separates the dishes that earn, the dishes that sell but carry too much cost, and the dishes that do neither. The card can tell you which dishes carry the most cost, which swing with the market, and which are cheap to produce but slow to sell. It cannot tell you what a guest will pay or what the hotel's reputation needs on the menu.

The decision options are few and concrete: re-engineer the recipe, re-price the dish, move it on the menu, or remove it. Each one needs a new card version, not a note in a drawer. There is no universal target cost percentage; a hotel sets its own from its own cost structure, and a figure borrowed from another property usually reflects a different building, a different wage bill and a different guest.

A menu change without a card change is how the two drift apart; treat the card as part of the menu, not as paperwork behind it. The price the guest pays and the fiscal document issued at the outlet are a separate matter, covered in the guide to fiscal devices in hotel outlets.

What to do next#

Start with one dish: the one you sell most, or the one you suspect. Write the card, run the yield test on its main ingredient, and cost it line by line. One honest card teaches the method faster than a project plan, and it shows you where the rest of the menu is guessing. From there the work is routine: one price list, dated card versions, and a monthly list of dishes whose cost moved.