Should I Cost Each Dish Separately or Use Flat Costing?

Use a mixed model, not one rule for the whole menu

I would not run my menu with only flat costing, and I would not force every dish into full individual costing either. In most restaurants, a mixed model is the sound answer: I cost separately when a dish carries real cost risk, and I use flat costing only where several items genuinely behave like one family. A veal cutlet, a fish special and a seasonal mushroom plate do not belong in the same shortcut logic as three focaccias built on the same dough, sauce base and side salad. The useful choice is practical, not ideological.

What matters to me is not which method sounds smarter on paper, but which one keeps prices believable when service is busy and suppliers move. Flat costing saves setup time, but it can hide which dish truly earns money and which only creates turnover. Full individual costing sounds precise, but it fails if nobody keeps recipes, yields and purchase prices up to date. So I decide by impact: anything that changes often, sells in volume, or affects margin noticeably gets its own calculation. Truly similar items can stay in a controlled group.

When I always cost a dish individually

I switch to individual costing when several warning signs appear together: high ingredient cost, unstable purchasing prices, a special prep path, relevant trimming loss, or sales volume that magnifies small mistakes. Think of a hypothetical bistro menu with veal schnitzel, pike-perch fillet, burrata salad and truffle pasta. They look like four mains, but economically they behave nothing alike. The schnitzel depends on meat cost, breading, frying fat and sides. The fish depends heavily on fillet size and yield. The salad moves with seasonal tomato prices. The pasta can become expensive through cheese, truffle product or finishing fat.

In those cases I do not just look at the headline ingredient. I cost the real sold portion, including base sauces, oils, herbs, garnish, bread, sides, trim loss and prep waste. That is exactly where seemingly similar plates begin to separate. If I keep recipes and quantities maintained, I can assign ingredients and amounts to products and compare recipe cost with my target food cost and selling price. That is why structured recipe data matters more than memory. It helps me understand what goes into every dish on your menu and base price decisions on maintained figures instead of late-night estimates.

When flat costing is clean, fair and useful

Flat costing makes sense only where my offers truly form one family, not merely where the names sound related. I use it when the cost structure is close enough that an average remains honest. A café-style group of filled focaccias is a good example: same dough base, similar sauce quantity, standard side salad, comparable topping weights, and only moderate differences between vegetable, ham or mozzarella versions. If the cheapest and most expensive variant stay in a narrow range, I can work with a calculated midpoint and keep the selling logic simple without distorting the economics.

Inside the team I use one hard rule: items stay in a flat-cost group only if they are operationally similar and commercially close. Three pasta dishes with the same pasta weight, the same base sauce and only minor topping changes may fit. The moment one version adds prawns, a costly cheese finish, extra prep stages or meaningfully different portion behaviour, it leaves the group. I also watch guest modifications. If one focaccia regularly gets double cheese, a substitution or an added side, grouping may stop being useful. Simplification should save work, not hide differences that matter.

A practical example with twelve main dishes

Take a hypothetical bistro with twelve mains: four strong lunch bestsellers and eight slower evening plates. I would not group the four bestsellers just because they sell often. High sales volume makes detailed control more important, because every small costing error repeats again and again. If a bowl, curry, burger or baked potato sells 180 times in a month, a modest over-portion on rice, avocado, meat or sauce has a much bigger effect than the same slip on a dish that sells fifteen times. Those items need target quantities for each component so I can see whether the issue is recipe, plating habit, side dish or price.

I judge quieter evening dishes differently. Two tarte flambée variants and two vegetarian oven dishes might be grouped if dough, cheese base, prep path and finish are genuinely close. But I would never force duck, a seasonal fish dish and a steak into that same simplification. There the raw material price, cooking loss and guest expectation are too important. So the real decision is not twelve times individual or twelve times flat. I ask which dishes carry margin, which fluctuate in purchasing, and which are alike enough that a family logic remains truthful. That keeps price reviews explainable to kitchen and service alike.

The criteria I use so the decision is not just instinct

To stop this becoming a mood-based debate, I use a short checklist. Is the main ingredient price-sensitive or seasonally unstable? Do portion size or yield differ meaningfully between variants? Are there relevant sub-recipes behind the dish, such as stock, sauce, dough or pickled components with their own yield? Does the dish sell enough that small errors matter over the month? Does it trigger frequent modifications that change cost or prep effort? If several answers are yes, I cost it individually. If almost none apply and the items are technically close, I test a flat-cost group instead.

I turn that into a small process instead of a one-off Excel exercise. When a new menu or seasonal change is coming, the kitchen lead reviews recipes and quantities, purchasing adds the current supply prices, and I assign each item either to individual costing or to a group calculation. After that, every sensitive item gets a fixed review point, for example when a supplier changes, when margin jumps become visible, or before the next menu swap. That creates more calm in the team because responsibilities are clear. At the same time, I avoid a common mistake: prices are calculated once with a lot of effort and then left untouched for months, even though portion size, purchasing or sales behaviour have long since shifted. For me, costing is therefore a recurring operating routine, not a one-time office project.

The common mistakes in both approaches

The biggest mistake in flat costing is laziness disguised as efficiency. I see five similar dishes and pretend their financial differences do not matter. That may look harmless until a supplier changes, one side dish gets upgraded, or one version suddenly becomes the bestseller. Then the average starts hiding a real margin problem. The biggest mistake in individual costing is the opposite. I build highly detailed item calculations, then nobody maintains them. The result is fake precision: the sheet looks exact while decisions are already based on outdated recipes and old purchase prices. I avoid both traps with priorities and review intervals.

Another trap is confusing a decent food-cost percentage with a good pricing decision. A dish can look mathematically acceptable and still be weak for the business if it consumes too much prep time, clogs the pass or anchors guest expectations too low. The reverse can also be true: a dish with relatively high food cost may still be strategically valuable if it attracts guests and drives drinks, starters or desserts. So I never judge from one number alone. I look at contribution margin, labour intensity, menu role and sales behaviour together. It helps to understand the day's activity before you close up so pricing discussions stay tied to what actually sold.

The practical answer for most restaurants

For most restaurants, tiered costing is the practical answer. I keep commercially important and risky dishes on individual calculations, group clearly similar items into defined flat-cost families, and manage very simple standard items with lighter maintenance. That is not an official industry rule; it is my organisational framework so the menu stays commercially useful and realistically maintainable. A small lunch restaurant with seven dishes may end up mostly on individual costing because each item matters. A café with many similar cakes, quiches and filled breads may group more strongly. What matters is that I know my exceptions and can justify them immediately.

So my answer is clear: cost dishes separately when they are expensive, volatile, high-volume or portion-sensitive, and use flat costing only for real product families with a tight cost structure. That protects margin without drowning the team in administration. I organise it fairly by treating recipes, quantities, yield and waste as shared operating data, not as private kitchen knowledge. Purchasing, kitchen, service and management work from the same logic, even if each handles a different part. For guests, that usually means fewer abrupt price swings. For employees, it removes confusion about why two similar-looking dishes are managed differently behind the scenes.