A Chef Said His Steak Had a 75.7% GP. It Was Really 70.9%.
This is Part One of a four-part series on where food margin actually goes wrong, starting with the most basic failure of all: getting the target wrong before anybody even walks into the kitchen.
A chef sent over his dish costs and GP calculations for a menu engineering report I was putting together. One steak dish was showing a GP of 75.7%.
That immediately raised an eyebrow. For the selling price involved, the margin looked unusually strong.
When I checked how he'd worked it out, I found the problem straight away: he'd left VAT in the selling price. Removing the VAT element brought the revenue down from £24.00 to £20.00 ex-VAT and the GP down to 70.9%.
Still a healthy margin. But nearly five percentage points lower than he'd calculated.
For those who live and breathe numbers, that might seem unbelievable. Yet it's a mistake I've seen more often than you'd expect.
Nobody stole anything, over-portioned or cut any corners - The number was wrong before anybody put their whites on! It was sitting in a spreadsheet, and every stocktake, variance report and performance review that followed was measuring results against a target that was never real from the outset.
That's what this series is really about.
Not theft or waste. Not poor portion control, though we'll come to those later.
This first part is about the two most ordinary, least dramatic ways a margin target gets set incorrectly from day one, making it impossible to achieve before service has even started.
You're A Tax Collector, Not The Owner Of That VAT
Here's the mental model that fixes this problem for good:
The moment money passes through your till, part of it isn't yours.
You're simply holding it on behalf of HMRC.
Any GP calculation that leaves VAT inside the selling price is measuring performance against revenue that was never yours to begin with.
Let's cost the steak properly:
Sale price: £24.00 incl. VAT → £20.00 excl. VAT
Cost: £5.83
GP calculated incorrectly (incl. VAT): 75.7%
GP calculated correctly (excl. VAT): 70.9%
Difference: 4.8 percentage points
Suddenly a dish that looked comfortably inside target has moved much closer to the area that deserves investigation. Nothing changed in the kitchen. The steak didn't become more expensive. The portion size stayed exactly the same.
The only thing that changed was removing VAT from a revenue figure.
For standard-rated food and drink sales in pubs and restaurants, VAT should be removed before GP calculations are made. If it isn't, every margin target that follows is immediately overstated.
The Other Fiction: Costing At A Desk, Not In The Kitchen
I've seen dish specifications built entirely from assumptions, costed at a desk, entered into a spreadsheet and then left untouched. That's a recipe for disappointment.
Take a joint of topside...
During cooking it's perfectly normal to lose around 25-30% of its weight through moisture loss and fat rendering. If your costing is based purely on invoice weight, without understanding what actually happens during cooking, you're almost certainly working with a number that doesn't reflect reality.
Let's assume:
A 9kg topside joint costs £12.00/kg
Total cost of the joint = £108.00
Cooking shrinkage is 30%
After cooking, the 9kg joint no longer weighs 9kg.
9kg × 70% = 6.3kg cooked weight
The important point is that the £108 cost hasn't shrunk. You still paid the same amount for the joint.
So the true cooked cost becomes:
£108 ÷ 6.3kg = £17.14 per cooked kg
Now look at a 200g portion:
Cost using raw weight: 200g × £12.00/kg = £2.40
Cost using cooked weight: 200g × £17.14/kg = £3.43
That's a difference of £1.03 per portion, or 43% higher than the original calculation.
Put that onto a roast dinner selling at £18.50 (£15.42 ex-VAT) and that £1.03 comes straight off gross profit.
Every additional £1.03 of food cost reduces GP by 6.68 percentage points.
So a dish an operator believes is achieving a 70% GP is actually running closer to 63.3% GP, before anyone has looked at vegetables, gravy, waste, or any other ingredient on the plate.
The mistake many operators make is costing meat using its purchase weight.
Guests are served roast meat after cooking, when shrinkage has already occurred. The cost of the whole joint needs to be recovered from the cooked weight that's left, not the raw weight that went into the oven.
9kg → 6.3kg → £17.14/kg → £3.43 per 200g portion
That's the simple calculation most people miss.
Different Problems. Same Mistake.
At first glance, the VAT example and the yield example appear unrelated.
One sits in the office, the other sits in the kitchen. In reality, they're exactly the same mistake as both involve accepting an erroneous figure that's written in a spreadsheet.
Once that assumption goes unchallenged, every stocktake, variance report, menu review and margin discussion that follows starts from the wrong baseline.
The business isn't measuring reality. It's measuring performance against a number that was always wrong.
Build It The Way You Cook It
The solution is remarkably simple.
Weigh the inputs. Cost them at today's invoice price.
Make the batch. Weigh what actually comes out.
Divide by the number of portions it genuinely serves.
A béchamel , a stock, a sauce reduction, a joint of beef - The method doesn't change.
Beef loses weight, rice and pasta gain weight as they absorb water, sauces reduce and concentrate.
The direction doesn't matter, the principle stays the same.
Cost what goes in. Measure what comes out.
Divide by what it actually serves. That's the only costing grounded in reality.
Not last year's incorrect spreadsheet.
Neither Of These Is A Kitchen Problem
A kitchen that's doing everything operationally correct can end up spending weeks investigating a 3% variance against theoretical food cost when the theoretical figure was wrong from the start.
A VAT error, an incorrect yield assumption, a recipe cost never validated in production.
Any one of them can create a gap that no amount of operational discipline will ever close.
What's potentially worse is that teams under pressure often find a way to "hit" an impossible target - Portions get shaved slightly, garnishes become smaller, a few grams disappear from every plate.
The numbers begin to improve, but only because people have adapted to a target that was never realistic in the first place. The business congratulates itself for fixing a problem that never really existed.
The bottom line is you can't hit a margin target that was fiction from the day it was set. Before looking for where the money's going, make sure the number you're aiming for is real.
I'm David Holden. I've spent 30 years working across hospitality audit and operations, helping hotels, pubs, restaurants and bars identify where profit is being lost and put practical controls in place to stop it.
How confident are you that your margin targets are based on reality?
If VAT, product yields, recipe costs or purchasing assumptions haven't been reviewed recently, there may be profit leaking from your operation before anybody picks up a pan in anger. I help hospitality operators identify where margin is being lost, separate fact from assumption, and recover profit that's already within the business.
Next in the series: Supplier decisions that quietly cost more than they save, and the simple purchasing habits that put margin back where it belongs.
Comments