How to calculate COGS for a restaurant
Cost of goods sold tells you what your kitchen really consumed, not what your recipes say it should have. The formula, a worked month, and the gap.
Cost of goods sold is what the food and drink you actually sold in a period cost you to buy. It is a period figure, not a dish figure, and it is the number your accountant, your bank, and your own profit and loss statement all care about.
It answers a different question from per-dish costing. A recipe tells you what one plate should cost. COGS tells you what your kitchen really consumed, including the waste, the over-portioning and the theft that a recipe card knows nothing about. You need both, and the gap between them is where the money leaks.
The formula
COGS = beginning inventory plus purchases minus ending inventory
Count what you have at the start of the period, add everything you bought during it, then subtract what is still on the shelf at the end. What is left is what you used.
A worked example
A cafe starts March with $4,200 of stock. It buys $11,800 of food and drink during the month. On March 31 it counts $3,600 still on hand.
COGS = 4,200 plus 11,800 minus 3,600 = $12,400
If that cafe took $41,000 in sales over the same month, its actual food cost is 12,400 divided by 41,000 = 30.2%.
Actual food cost versus ideal food cost
The 30.2% above is your actual food cost. Your ideal food cost is what the same sales should have cost if every dish had been made exactly to recipe: multiply each item's recipe cost by how many you sold, and divide by revenue.
The difference between the two is your variance, and it is the most useful number in this whole exercise. An ideal of 27% against an actual of 30.2% means 3.2 points of revenue left the building without being sold. On $41,000 that is about $1,300 in one month.
Variance is normal. Zero variance usually means someone is guessing at the count. Consistently more than 2 to 3 points is worth investigating.
Where the variance usually comes from
- Over-portioning. The most common cause by far, and the least visible.
- Waste and spoilage that never gets recorded.
- Recipes that are out of date, so the ideal figure is wrong rather than the actual.
- Stale ingredient prices. If your supplier raised chicken 12% and your recipes still carry the old cost, your ideal food cost is fiction.
- Staff meals and comps counted as sales rather than as cost.
- Theft, which is real but is usually smaller than the first four.
Counting inventory without hating it
Count the same items, in the same order, at the same time of the period, using the same units you buy in. Consistency matters more than precision: an inventory that is 3% off every month still gives you a trustworthy variance trend, while one that is perfect in March and rushed in April tells you nothing.
Count at the close of the last day of the period, before any delivery for the next one lands.
What COGS does not include
Labor is not in COGS. Neither is rent, utilities, marketing, or equipment. COGS is ingredients, drink, and packaging only. Add labor to it and you have prime cost, which is the metric most operators actually manage the business by.
Frequently asked questions
- How often should I calculate COGS?
- Monthly is enough for most operators, and it lines up with the period your profit and loss statement already uses. Weekly counts are worth it if your variance is high and you are actively hunting the cause, but they are a real time cost.
- Does COGS include packaging?
- Yes. Anything that leaves with the order was bought and consumed in the period. Note that Dishboard keeps packaging separate from a dish's recipe cost so it can be applied per channel, so a per-dish figure and this period figure are deliberately scoped differently.
- Why is my actual food cost higher than my recipe costs suggest?
- That gap is your variance, and it is the point of calculating COGS at all. Over-portioning, unrecorded waste, and stale ingredient prices are the usual causes, roughly in that order.
- Can I calculate COGS without counting inventory?
- No. Purchases alone tell you what you bought, not what you used, and the difference is exactly the number you are trying to find. A month with a big delivery on the last day would look catastrophic and mean nothing.
Put in your opening stock, purchases, closing stock and sales, and see the variance the two figures hide.
Free, and no account needed.
Free tools
Put the math to work in seconds. No signup required.