Whole Fish or Ready-Cut Fillets: A Yield and Costing Decision
A seaside taverna serves a grilled sea bream fillet at €9.80 including 9% VAT, and sells about 50 on a Saturday. The fishmonger quotes whole wild sea bream at €8.40/kg. Last month's yield test on the same species, done on 12 kg, gave 62% skin-on fillet yield after removing head, frame, fins and belly trim, and the chef believes it is repeatable within one or two points. As an alternative the supplier offers ready-cut, pin-boned fillets at €16.90/kg, with about 4% trim loss on arrival. A commis takes roughly 20 minutes to fillet 12 kg of whole fish and is paid €14 per hour. Frames and heads currently go into the bin. The owner's food-cost target is 30% of the pre-VAT selling price, calculated on the full plate, and the garnish, olive oil, lemon and potatoes add about €0.70 per portion.
1. Calculate the food cost of a 180 g raw fillet portion both ways, including labour, and say which choice is cheaper per portion and by how much.
Read the worked answer
Convert everything to purchase weight, then to money.
Whole fish. A 180 g fillet is 62% of the whole fish, so purchase weight = 180 ÷ 0.62 = 290 g (0.290 kg). Cost = 0.290 × €8.40 = €2.44 per portion. Labour: 12 kg whole yields 7.44 kg fillet, which is 7.44 ÷ 0.180 = 41 portions. Twenty minutes at €14/hour = €4.67, so €4.67 ÷ 41 = €0.11 per portion. Total ≈ €2.55.
Ready-cut fillets. 4% trim loss means you buy 180 ÷ 0.96 = 187.5 g to plate 180 g. Cost = 0.1875 × €16.90 = €3.17 per portion, with no kitchen labour.
Whole fish is cheaper by about €0.62 per portion, roughly €31 across the 50 Saturday covers, plus the value of the frames. That is a real saving, but note that it is only about 4% of the selling price: the decision should not be made on this number alone. The ready-cut option also removes a 20-minute bottleneck at the busiest point of the day and gives a more consistent portion, so if the commis is the constraint on a Saturday, paying €0.62 is a defensible purchase of time. Show the arithmetic and let the owner choose with the trade-off in front of them, rather than presenting the cheaper option as automatically correct.
2. Test the current €9.80 menu price against the 30% target, using the full plate cost rather than the protein alone. What do you advise?
Read the worked answer
Strip the VAT first, because food-cost percentage must be measured against the net selling price: €9.80 ÷ 1.09 = €8.99 net.
Protein only: €2.55 ÷ €8.99 = 28.4%, which looks comfortably inside target. But the target is defined on the whole plate, and the garnish, oil, lemon and potatoes add €0.70. Full plate cost = 2.55 + 0.70 = €3.25, so the true food cost is 3.25 ÷ 8.99 = 36.1%. The dish is over target by about six points, roughly €0.55 of margin per portion.
The ready-cut version makes it worse: (3.17 + 0.70) ÷ 8.99 = 43.0%. So the choice of fish form is not the problem to attack first, and switching to ready-cut fillets on the grounds of convenience would deepen an existing loss.
Advice, in order of preference: (1) re-specify the garnish, since €0.70 on a €9.80 dish is 8% of the selling price and is usually where the slack is; (2) raise the price to about €10.80, which brings the net to €9.91 and the cost to 32.8%, still under the ceiling of what a seaside taverna can charge; (3) only then decide on whole versus ready-cut. Also use the frames and heads for stock or fish soup rather than binning them. A food-cost target is a ceiling, not a price-setting formula, but a dish that sits six points above it is subsidising the rest of the menu.
3. The chef proposes using the frames and heads to make a fish stock for the kitchen. How does that change the costing, and how should it be reflected on paper?
Read the worked answer
It changes the costing through the yield test, not through the menu price. The €2.44 protein cost was calculated on the basis that head, frame and trim are waste. If they are converted into stock, part of the purchase price is recovered by another dish, so the effective cost of the fillet falls and the true cost of the stock is close to zero apart from labour and vegetables.
On paper: (1) record the yield test properly - 12 kg in, 7.44 kg fillet, 2.1 kg frames and heads, 2.46 kg true waste - and keep the sheet on file so the 62% figure is auditable; (2) transfer the frames to stock at a nominal value, or cost the stock as fennel, onion, wine and labour only, so you do not double-charge the fish to two dishes; (3) use a standard recipe for the stock with a known yield in litres so that a portion of fish soup has a defensible cost.
The reasoning matters more than the accounting trick: by-products only have value if there is a use for them at the volume produced. Fifty portions on Saturday produce about 5 kg of frames, which is more than a small kitchen can turn into stock in a week unless it is reduced and frozen. So the honest answer is that the frames reduce waste and improve the soup's cost, but they do not rescue a dish that is 36% food cost; price and garnish still have to be dealt with.