Written by: JJ Tan, Founder, Jelly
Key Takeaways For UK Kitchens
- A 4% gap between theoretical and actual food cost on £40,000 of monthly sales creates a £1,600 loss that points to specific operational failures.
- The main drivers of variance are over-portioning, yield loss, spoilage and waste, unrecorded staff meals, supplier price changes, and inventory counting errors.
- Weekly stock counts and reconciliations against POS sales keep variance below 2%. Monthly counts allow issues to grow across 20 or more services before anyone notices.
- UK operators face rising cost pressure, with food price inflation forecast to reach 3.9% by December 2026 and 6.4% by July 2027, so small variances now decide profit or loss.
- Jelly automates invoice scanning, price alerts, and live menu costing so the variance closes in the background. See how Jelly works in your own kitchen.
Theoretical Vs Actual Food Cost: Clear Definitions
Theoretical food cost is what your menu should cost based on recipes, portion sizes and current ingredient prices, multiplied by the units sold. Actual food cost is what your kitchen really spent, calculated from opening stock plus purchases minus closing stock over the same period.
| Basis Of Calculation | Data Source | What It Tells You | What A Gap Signals |
|---|---|---|---|
| Recipe specifications × units sold | Recipe cards, POS sales mix | What the menu should have cost | Operational failure between recipe and plate |
| Opening stock + purchases − closing stock | Invoices, stocktake counts | What the kitchen actually spent | Waste, over-portioning, price creep, or counting error |
A note for UK operators: supplier invoices arrive VAT-inclusive. The correct method for removing 20% standard-rate VAT is to divide the gross total by 1.20, not to subtract 20% from it. A £360 VAT-inclusive invoice is £300 net, not £288. Food cost calculations should always use the net (ex-VAT) figure. Applying the wrong formula on every supplier invoice distorts actual food cost in a meaningful way.
The Formulas And A UK £ Worked Example
Use these two formulas every time you calculate food cost:
- Theoretical food cost % = (Total recipe cost of dishes sold ÷ Food sales) × 100
- Actual food cost % = ((Opening stock + Purchases − Closing stock) ÷ Food sales) × 100
Running the same numbers through both formulas for a London gastropub in a four-week period:
- Food sales: £40,000
- Opening stock: £8,000
- Purchases: £14,600
- Closing stock: £9,000
- Total recipe cost of dishes sold: £12,000
Theoretical food cost: £12,000 ÷ £40,000 × 100 = 30.0%
Actual food cost: (£8,000 + £14,600 − £9,000) ÷ £40,000 × 100 = £13,600 ÷ £40,000 × 100 = 34.0%
Variance: £1,600 or 4.0 percentage points.
State the variance in both figures. The percentage lets you benchmark against previous periods. The £ figure shows the cash impact this month.
Industry guidance treats a food cost variance under 2% as well-run and anything past 5% as a systemic leak. A variance of 5% or more points to a systemic problem. These are widely used industry benchmarks. The 4.0% variance in this example sits in the warning zone and needs investigation now.
The Variance Bridge: From Gap To Named Causes
The variance bridge is actual food cost minus theoretical food cost, expressed in both £ and as a percentage of food sales. It turns one blunt number into a list of specific causes you can act on.
The £1,600 gap in the worked example rarely comes from a single failure. A realistic attribution across a gastropub might look like this:
- Over-portioning in the best-selling dishes. At the case-study trattoria, over-portioning concentrated in the three best-selling dishes drained the equivalent of 3.8 EBITDA points every month.
- Yield loss on butchery and fish filleting.
- Spoilage and waste.
- Unrecorded staff meals and comps. Unrecorded comps and staff meals account for 0.3–0.8% of food cost variance, which on €80K monthly revenue equates to roughly €240–€640 per month.
- Supplier price creep that has not been updated in recipes.
Each of those lines needs a different operational fix. A 4% aggregate food cost variance can mask proteins running at 10% while dry goods sit at 0.5%. The blended number hides the items destroying margin. The bridge makes them visible.
Food cost variance has four primary causes by frequency: portioning inconsistency (42% of variance), untracked waste (31%), shrinkage and theft (18%), and system data errors (9%). Knowing that split tells you where to look first.
What Causes A Food Cost Variance?
The gap you saw in the bridge usually breaks into a small set of recurring issues. Each one has a clear fix.
- Over-portioning: Portions drift above spec on busy services. Plating 7 oz of protein against a 6 oz recipe specification makes food cost run 17% higher than theoretical cost. Fix this with portion scales, standardised serving utensils and spot checks against the recipe card.
- Yield and trim loss: The recipe assumes a yield the kitchen is not achieving on butchery, fish filleting or prep. Yield testing is critical to theoretical food cost because it accounts for weight lost during trimming and cooking. If only 70% of a whole side of beef is usable for steaks, theoretical cost must be based on the edible portion price, not the as-purchased price. Fix this by recording actual yields and updating yield factors in the recipe cost.
- Spoilage and waste: Over-ordering, poor rotation and short shelf life all push cost up. 4% to 10% of food purchased becomes pre-consumer waste. Fix this with par levels tied to covers, date labelling and a waste log that feeds back into ordering.
- Unrecorded staff meals and comps: Food leaves the kitchen without hitting the till. Ringing staff meals through the POS as a £0 or discounted item means inventory is deducted from theoretical usage; otherwise staff meals appear as unexplained leakage in the variance report. Fix this with a simple staff meal and comps log reconciled against POS sales.
- Supplier price creep: The invoice price has moved but the recipe cost has not. Across operators, the average number of ingredient price changes per month exceeds forty for a mid-size restaurant group, making static recipe costs unworkable. Fix this with line-item invoice checking and price alerts so increases are caught the week they happen.
- Inventory counting error: Miscounts at stocktake, especially on high-value items. A stocktake that is wrong by 3% in either direction can swing a restaurant’s COGS percentage by a full point. That swing is enough to hide a real variance or create a false one. Fix this with consistent count sheets, blind counts and a second pair of eyes on the top 20 lines by value.
Standard Cost, Actual Cost, And Inventory Variance
Standard cost vs actual cost: Standard cost (also called theoretical cost) is the planned cost per unit built from the recipe and expected yield. It reflects what a dish should cost if every ingredient is measured correctly and every yield assumption holds. Actual cost is an amount determined on the basis of costs incurred, including standard cost properly adjusted for applicable variance. The operational efficiency ratio measures adherence to cost standards. It is calculated by dividing actual cost percentage by standard cost percentage. A ratio close to 1.0 means costs are on target. A ratio above 1.0 means real costs are exceeding targets. The gap between standard and actual cost is the same variance viewed from the cost side rather than the sales side, and the causes match the variance bridge list.
Actual vs theoretical inventory: Theoretical inventory is what the system says you should have on hand given purchases and recorded sales. It equals opening stock plus purchases minus what the recipes say should have been consumed. Actual inventory is what the stocktake counts. Variance is defined as actual usage minus theoretical usage. A positive number means more was used than sold, which signals a cost problem with a cause. A negative number usually indicates a counting problem. The difference between theoretical and actual inventory is usage variance, which is the same operational failures from the cause list above showing up in units rather than pounds. A kitchen running unaccounted chicken each week is showing the same portion-drift problem as that 4% food-cost variance, since a 1.5-oz over-pour on an 8-oz chicken breast across 400 covers per day adds up to 37.5 lbs of unaccounted chicken per week. The inventory view simply makes it easier to see which ingredient is driving it.
What To Do This Week In Your Kitchen
For a UK operator who has just found a variance, start with the fastest wins and move to deeper fixes.
- Check the top 20 invoice lines for price increases by comparing this week’s net (ex-VAT) prices against last period’s. Price creep moves quickly and is easy to confirm.
- Recount high-value stock lines such as proteins, fish and dairy with a second person present. Counting errors distort every other number.
- Audit portioning on the five best-selling dishes by weighing plates during a service. This is where most over-portioning hides.
- Start a staff meal and comps log and reconcile it against POS voids and discounts. This closes a common blind spot.
- Update yield factors on butchery and fish using actual trim weights from this week’s prep. That keeps theoretical cost honest.
Operators who count stock weekly and reconcile against POS sales consistently achieve food cost variance below 2%, while operators who count monthly discover problems 20 or more service shifts too late. The checklist above is a one-week intervention. The permanent system is what closes the gap automatically.
The UK cost environment makes this urgent. The Food and Drink Federation forecast on 9 September 2026 that UK food and non-alcoholic drink price inflation would reach 3.9% by December 2026, before peaking at 6.4% in July 2027. Drivers include the ongoing impact of the U.S.-Iran war, drought in Europe and the El Niño weather pattern. The ONS reported that the CPIH ‘Restaurants and hotels’ division rose 4.0% in the 12 months to July 2026. Hospitality output prices are rising roughly three times faster than food input prices, which compresses the margin between what operators charge and what they pay. Industry data shows that 23% of UK hospitality businesses are now operating at a loss, up from 15% three months earlier. In that environment, the £1,600 gap identified earlier is the difference between profit and loss. Closing it requires the three disciplines above to run continuously, which is where automation earns its place.
How Jelly Closes The Gap For UK Operators
Jelly gives growing UK restaurants, pubs and boutique hotels a simple way to manage food and beverage costs by automating invoices, inventory and real-time menu profitability.
Automated invoice scanning digitises every line item from a photo or email. It captures quantity, SKU, price and tax, so supplier price creep is caught the week it happens rather than at month-end. The Price Alert feature flags every ingredient price increase or decrease, by how much and from which supplier, giving hard data for supplier negotiations and credit notes. As Stuart Noble, Head Chef at Cairn Lodge Hotel, put it: “Price hikes were crushing our margins. I felt helpless. With Jelly, every dish cost is up-to-date at my fingertips. We slashed food costs by 5% in a month.”
Live dish costing in the Kitchen section keeps recipe costs and GP margins current with every new invoice, so the theoretical figure stays accurate without spreadsheet maintenance. What used to take 28 minutes to cost a single menu item now takes 3 minutes. The Flash Report gives a daily, weekly or monthly view of gross profit margin from invoice costs and POS sales. Variance becomes visible in real time rather than weeks later.
POS integrations with Square, EPOS Now, Toast and Lightspeed deliver item-level sales data mapped to Jelly dishes. Setup takes around five minutes across all four systems. Xero integration pushes digitised invoices straight into accounting. Jelly works alongside these tools as the control layer for cost and margin.
Jelly charges a flat £129 per location per month. Customers cut food costs by 3% on average in the first three months and save 10–20 hours of admin every month.
Watch a short demo of Jelly’s variance bridge using your own numbers.
The Gap Is A Diagnostic, Not A Judgment
The gap between theoretical and actual food cost is a diagnostic that names the operational failure. It is not a judgment on your kitchen. That 4% variance on £40,000 of food sales equals £1,600 a month, and it also points to over-portioning on a handful of dishes, yield loss on butchery, spoilage from over-ordering, unrecorded staff meals and price creep from a supplier who updated their list weeks ago. Each of those has a fix. None of them is visible without the comparison.
The fix path stays simple: accurate recipe costing, line-item invoice visibility and a weekly variance review. The difficulty lies in keeping all three current at once. Recipe costs drift the moment a supplier changes a price. Invoice data sits in paper files until someone enters it. Weekly reviews get skipped when service is busy. Jelly automates the first two so the third becomes a five-minute read rather than a two-hour reconciliation.
Ruth Seggie, Owner of The Howard Arms, put it plainly: “Our accountant said we’d be lucky to hit 60% gross profit. After using Jelly, we reached 80%. Now I sleep better knowing my costs are under control and can react instantly, not weeks later.”
If you want that same control over your own numbers, we can show you how it works.