Skip to content

Part 2 of 7 · Menu cost calculator series ~5 min read

How yield turns purchase price into plate cost

Yield is the largest single correction between an invoice and a plate, it varies by supplier and by season, and it is almost always taken from a table somebody found rather than from a scale.

Key takeaways

  • Measure yield in your own kitchen, three times, and average it.
  • Yield varies by supplier and by season; re-measure when either changes.
  • Trim loss and cooking loss compound; apply both.
  • Trim with a use has a value and should be credited, with the assumption stated.
  • A yield below expectation is a supplier finding, not just a costing input.

Measuring it

How an ingredient yield is measured in the kitchenA vertical chain of five steps entered by a box labelled A delivery arrives, weigh it as received. Step one weighs gross, before anything. Step two prepares it as normal, by the same person in the same way. Step three weighs the usable portion that reaches the line. Step four weighs usable trim such as bones and offcuts for stock, with a side box saying it is credited at a stated value. Step five records the yield with a date, repeating three times. A note says three measurements take twenty minutes in total and it is the highest-value data in the system.AWS ACCOUNTA delivery arrivesweigh it as receivedWeigh grossbefore anythingPrepare as normalthe same person, the same wayWeigh usablewhat reaches the lineWeigh usable trimbones, offcuts for stockCreditedat a stated valueYield, and a daterepeat three timesThree measurements, twenty minutes total. It is the highest-value data in the system.
Fig 1. How a yield is established. It is a kitchen exercise rather than a data exercise, and it only has to be done once per ingredient per supplier.
  • Compute
  • Machine learning
  • Analytics
  • Front-end & mobile
  • People

Why not a published table

Published yields describe an average specification prepared by an average person. Your supplier’s trim specification, your chef’s preferences and your portioning all differ, and the spread between kitchens on the same ingredient is wide enough to move a dish cost by twenty per cent.

The table is a reasonable starting point for a dish that has never been made. It should be replaced by a measurement the first time the ingredient is actually prepared, and flagged as unmeasured until then so nobody mistakes it for a fact.

It changes

A different supplier, a different specification, a seasonal change in size — all move yield. Small root vegetables have worse yield than large ones because peeling loss is proportional to surface area, which is why winter costings on some vegetables are genuinely different from summer ones.

Re-measuring quarterly on the top ten ingredients by spend covers most of the exposure and is an hour a quarter.

Two losses, compounded

How trim loss and cooking loss compound on one kilogram of an ingredientA stacked bar chart with three bars in grams. Two series: grams reaching the plate in green, and grams lost in red. Purchased: one thousand grams. After trim at sixty-eight per cent: six hundred and eighty usable, three hundred and twenty lost. After cooking at eighty-two per cent: five hundred and fifty-eight usable, four hundred and forty-two lost. A note says a kilo bought becomes five hundred and fifty-eight grams served, so the effective cost is seventy-nine per cent higher rather than forty-seven.0500100015002000~1000Purchased~1000After trim (68%)~1000After cooking (82%)Grams reaching the plateGrams lostA kilo bought becomes 558g served. The effective cost is 79% higher, not 47%.
Fig 2. Trim loss and cooking loss applied in sequence. Applying only one of them is the common error and it understates the cost substantially.

Which of the two applies depends on how the recipe states its quantities. A recipe written in cooked weight on the plate needs both; one written in prepared raw weight needs only the trim yield.

That distinction has to be explicit per recipe line, because getting it wrong in either direction produces a cost that is out by twenty per cent and looks entirely plausible.

Trim that has value

How to treat the offcuts

  • Bones and trim that become stock have a real value: what you would otherwise buy.
  • Credit them at that replacement value, not at the purchase price of the original ingredient.
  • Only if they are actually used. Trim that is thrown away when the stock pot is full is a cost.
  • State the assumption on the costing, because it is a judgement and it can be generous.
  • Staff food is a cost, and belongs in its own line rather than hidden in yield.
  • When in doubt, do not credit it. An uncredited yield is conservative and an over-credited one silently improves every dish on the menu.

The last line is worth holding to. Yield credits are the easiest place for optimism to enter a costing model, and once it is there it improves every dish by a few per cent in a way nobody can see.

Yield as a supplier signal

A yield that drops from sixty-eight to sixty-one per cent on the same ingredient is a supplier telling you something: a change in specification, more fat, more trim, smaller units. The price per kilo may not have moved at all while the cost per plate went up eleven per cent.

That is a genuinely useful finding and it is invisible without measured yields. It is also a specific, evidenced conversation with the supplier rather than a general complaint about quality.

Next: why a costing needs a date.

All posts