Comparing actual and theoretical food cost sets the cost of the dishes you sold, according to their recipes, against the value of stock actually used. Theoretical = quantities sold × recipe cost. Actual = opening stock + purchases − closing stock. Both values must cover the same period, the same product scope and a comparable valuation. Purchases alone are not actual usage; how to derive the actual value from stock and purchases is worked through for a model business in calculate food cost percentage.
Contents
- Calculating theoretical and actual food cost
- Worked example: a €2,400 variance
- Per cent or percentage points?
- Check variances in the right order
- Separate daily signals from a reliable stock count
- FAQ
Calculating theoretical and actual food cost
Theoretical food cost = sum of quantity sold per dish × recipe cost per portion.
For this you need complete recipes, matching units and a mapping of the POS items sold. Changes to portions and purchase prices have to be taken into account. Be explicit about which valuation applies to the period; otherwise a price change can look like a usage variance.
Actual food cost = opening stock + purchases − closing stock.
Value opening and closing stock using a consistent method. Deliveries, returns and transfers belong in the correct period and the correct location. Compare food or beverages each against the matching product and sales scope.
Variance in euros = actual food cost − theoretical food cost.
The basics are explained under the food cost ratio. The food cost calculator helps you trace usage and percentage.
Worked example: a €2,400 variance
The following amounts are model assumptions for one restaurant month. They do not prove any saving through Kiosa.
| Item | Amount |
|---|---|
| Net sales | €80,000 |
| Theoretical food cost from recipes and sales | €21,600 |
| Opening stock | €8,000 |
| Purchases | €22,000 |
| Closing stock | €6,000 |
| Actual food cost: 8,000 + 22,000 − 6,000 | €24,000 |
| Variance: 24,000 − 21,600 | €2,400 |
The theoretical percentage is 27 per cent, the actual percentage 30 per cent. The gap is three percentage points of net sales. Relative to theoretical food cost, however, 2,400 ÷ 21,600 × 100 is 11.1 per cent excess usage or excess cost.
This variance is a task to investigate. It proves neither waste nor theft. Missing recipes, differing prices, incorrectly recorded stock or undocumented staff meals can explain it just as well.
Per cent or percentage points?
If the food cost percentage rises from 27 to 30 per cent, it rises by three percentage points. The relative change is 30 ÷ 27 − 1 = 11.1 per cent. Both statements are correct, but they answer different questions.
So label your reports with denominator and unit: euro variance, percentage points of net sales or per cent of theoretical food cost. Without that, product groups and locations can't be compared reliably.
Check variances in the right order
| Check | Typical source of error | Next step |
|---|---|---|
| Period and stock movements | delivery, return or transfer in the wrong month | reconcile documents and booking dates |
| Units and stock count | pieces confused with kilograms, pack size changed | check the count and conversion factors |
| Recipe and POS coverage | sale without an assigned recipe, outdated portion | add the missing mapping or recipe |
| Price valuation | theoretical at old prices, actual at new prices | align valuation and price date |
| Documented usage | staff meals, comps, breakage or spoilage missing | record movements and write-offs |
| Operational process | actual portion differs from the recipe | spot-check the kitchen process |
Check the data first, then the operation. If only part of your sales is covered by recipes, you must not treat that partial theoretical value as the full counterpart of total actual usage.
Separate daily signals from a reliable stock count
Recipe costs, sales, deliveries and price changes can provide daily signals. A reliable actual usage figure for a period, however, requires a matching opening and closing stock count. An automatically rolled-forward theoretical stock level is not an independently measured closing count.
So in your report, separate calculated expectation, recorded stock movements and counted stock. Set the counting rhythm by stock value, turnover and effort. For a single expensive or conspicuous product group, counting more often can make sense.
The Kiosa Inventory module connects the recipe, purchasing and stock count sides. The scope of an analysis depends on maintained recipe mappings and complete stock data. The Inventory module costs €49 per location per month on top of One View; One View costs €99. The supplier price analysis and menu pricing complement the search for causes.
From analysis to action
What this is about: the pillar for theoretical recipe cost, actual usage and the causes of variance.
Put it into practice: Actual vs. theoretical analysis in Kiosa Inventory
More on Food cost and inventory
- Food Cost Percentage Formula: Calculation, Example, Calculator
- Restaurant Shrinkage: Where the Percentage Points Disappear
- Restaurant Inventory Count: How Often, What to Count + Free Template
FAQ
How do I calculate theoretical food cost?
Multiply the quantity sold of each dish by its recipe cost per portion and add up for the period. Missing recipe mappings and outdated prices limit how meaningful the result is.
How do I calculate actual food cost?
Opening stock plus purchases minus closing stock, each valued consistently and for the same period. Returns and transfers must be accounted for correctly.
What theoretical-vs-actual variance is normal?
A general threshold isn't reliable without knowing the concept, data coverage and valuation method. Quantify the variance, compare similar periods and check the causes. A flat shrinkage rate is no substitute for documentation.
Can I compare purchases directly with recipe costs?
Not as a complete usage comparison. If your stock level rises or falls, that changes actual usage relative to purchases. Take opening and closing stock into account.
Is a daily analysis possible without a stock count?
As an operational estimate or early signal, yes; as independently measured actual usage, no. Make this distinction explicit so a calculated stock level doesn't turn into a supposedly exact stock count variance.
The stock count as the foundation of the comparison
The stock count guide with Excel template shows how to record opening and closing stock and purchases for the same period. The Excel file helps with stock valuation and a transparent usage calculation. It doesn't import POS data automatically and doesn't on its own prove the cause of a variance.
Your next steps
Choose a clearly defined product group and a period with an opening and closing stock count. Check stock movements, quantities sold and recipe coverage before comparing theoretical and actual. Use the food cost calculator and location benchmarking to put the percentage per location into context.
Want to see this process for your restaurants? See Kiosa in a demo call.