Skip to content
allanninal.dev/build
Collaborate

Part 5 of 7 · Menu cost calculator series ~5 min read

How to read a dish profitability report without getting it wrong

A dish profitability report is the most misread document in hospitality, and the two classic errors both come from looking at one number in isolation.

Key takeaways

  • Percentage and cash margin together, never one alone.
  • Multiply margin by units sold; that is what pays the rent.
  • Some dishes exist to bring people in or to complete a menu.
  • Change the recipe or the price before removing a dish.
  • A dish selling twice a week is a data question, not a profitability one.

The two-number rule

Coffee and steak compared by gross profit percentage and by weekly cash marginA bar chart with four bars. Two series: gross profit as a percentage in purple, and cash margin per week in pounds in green. Coffee: eighty-nine per cent gross profit. Steak: sixty-eight per cent. Coffee cash margin: two hundred and sixty-five pounds a week. Steak cash margin: one thousand three hundred and thirty pounds a week. A note says coffee wins on percentage while steak pays five times more of the rent.0500100015002000~89Coffee~68Steak~265Coffee, cash~1330Steak, cashGross profit, %Cash margin per week, £Coffee wins on percentage. Steak pays five times more of the rent.
Fig 1. Two items compared on percentage and on cash. A menu managed on percentage alone systematically promotes the wrong things.

The percentage is useful for comparing similar items and for spotting a dish whose cost has drifted. It is actively misleading as a menu management tool, because a business pays its rent in pounds rather than in percentages.

So every line carries both, and the report is sorted by total cash margin: margin per dish multiplied by how many sell. That ordering answers the question most people are actually asking.

Popularity and margin together

Four combinations of dish popularity and margin and what each needsThree boxes stacked on the left. Sells well with good margin, labelled leave alone: protect it. Sells well with poor margin, labelled the opportunity: fix the recipe or the price. Sells badly with good margin, labelled presentation: promote it or move it on the menu. All three converge on Sells badly with poor margin, the only removal candidate, and that leads down to And even then, check what it is for. A note says the second lane is where nearly all the money is and it is the least acted on.Sells well, good marginprotect itleave aloneSells well, poor marginfix the recipe or pricethe opportunitySells badly, good marginpromote it, or move itpresentationSells badly, poor marginthe only removal candidateAnd even thencheck what it is forThe second lane is where nearly all the money is, and it is the least acted on.
Fig 2. The four quadrants and what each one calls for. Only one of them is a candidate for removal, and even that one needs a question asked first.
  • Compute
  • Machine learning
  • Analytics

The second lane

A popular dish with a poor margin is the single largest opportunity on most menus, because a small change is multiplied by a large number of covers. Twenty pence off the cost of something that sells forty times a week is four hundred pounds a year from one adjustment.

The changes available are a smaller portion of the expensive component, a different supplier for it, a garnish that adds perceived value cheaply, or a price rise. All four are less drastic than removing a dish people evidently want.

Dishes that exist for other reasons

Before removing anything, ask what it is for

  • The signature dish that people come for and then bring three friends who order other things.
  • The menu completer: the vegetarian option, the children’s option, the thing that means a group of six can all eat here.
  • The anchor: the expensive item that makes everything else look reasonable, and which does not need to sell much.
  • The one that uses the trim from something else, and whose removal makes another dish more expensive.
  • The staff favourite is not a business reason, and it is worth saying so plainly.
  • None of these are visible in the numbers, which is why the report proposes nothing.

The fourth item is the one a costing system can actually help with, and it is worth checking automatically: removing a dish that consumes the trim from another ingredient changes the yield credit on that ingredient and makes the other dish more expensive. That interaction is invisible unless something is looking for it.

The dish that sells twice a week

Why dishes with few sales are excluded from profitability rankingsA horizontal row of five boxes. Two sales a week, eight in the month. Margin looks bad, on eight data points. Or looks great, equally likely. Neither is evidence, so say so. Grey it out, below twenty sales. A note says the most dramatic movements on any menu report are always the least-sold dishes.SMALL NUMBERS2 sales a week8 in the monthMargin looks badon 8 data pointsOr looks greatequally likelyNeither is evidencesay soGrey it outbelow 20 salesThe most dramatic movements on any menu report are always the least-sold dishes.
Fig 3. Why low-volume dishes are annotated rather than ranked. The same statistical caution as everywhere else in this series, applied to a menu.
  • Compute
  • Management
  • Analytics

Greying out or annotating low-volume lines stops the report from generating a monthly conversation about a dish whose numbers are noise, which is otherwise where a good deal of the attention goes.

What the system does not do

It does not recommend a price, it does not recommend removing anything, and it does not compute an optimal menu. Those all require knowing what dishes are for, what the competition charges, and what the room will bear, none of which are in the data.

What it does is make the arithmetic correct and current, decompose every change, and put the two numbers that matter next to each other. The decisions stay with whoever knows the restaurant.

Next: what all of this costs to run.

All posts