Before we started doing stock takes properly, we had a rough sense of what we were spending on raw materials each month. We knew the purchasing figure because it came from the invoices. What we did not know was whether what we bought matched what we sold, and what the difference meant.
A stock take answers that question. It is a physical count of every ingredient in the kitchen, bar, and storage, valued at the price you paid for it, done at a consistent point in the billing cycle. The difference between what your opening stock plus purchases should theoretically leave you with, and what you actually count, is the gap. That gap tells you a great deal about what is happening operationally.
How We Structure Ours
We run a stock take at the end of every month, after the last service on the final trading day. The timing matters. Doing it mid-month or mid-week means the count reflects a partial cycle, and the reconciliation becomes harder.
Two people count, independently if possible. One person counts and calls out quantities. The other records. When the count is complete, the sheet goes to whoever is reconciling the financials, which at our scale is usually me.
We organise the count by storage area, not by category. Kitchen cold room, kitchen dry store, bar fridge, bar dry store, and the prep area. Moving through physical space is faster and less error-prone than trying to hunt down every item by category across different locations.
What We Are Counting and Why
Every ingredient in the operation is counted, but not all of them require the same precision. High-value, high-consumption items like meat, dairy, and specialty coffee beans get a careful count. They represent a large proportion of the food cost and any variance there is significant.
Lower-value items like basic spices, sugar, and condiments get a reasonable count, not an obsessive one. The time spent counting fifty-gram discrepancies in a bag of cumin is not worth the information it produces.
Beverages, including wine and spirits if you serve them, should be counted at the bottle level. The cost per bottle is known, the opening and closing stock should reconcile against sales, and any variance in beverages is worth investigating carefully because the margins on drinks are usually higher and losses there are more visible.
What the Numbers Tell You
After the count, the reconciliation is straightforward in principle. Opening stock, plus purchases received during the month, minus what the sales data says you should have used, equals the theoretical closing stock. The actual count is the real closing stock. The difference is your variance.
A small variance, under two percent of total food cost, is normal. It reflects minor inaccuracies in measurement, small amounts of plate wastage, and the natural imprecision of running a kitchen.
A variance above three or four percent starts to indicate something specific. It might be prep wastage that is not being tracked — one of the most consistent sources of unexplained variance in small kitchens. It might be portion inconsistency that has grown over time. It might be something more direct.
In our experience, the category where the variance appears tells you where to look. If the variance is concentrated in meat, the kitchen’s portioning is where you start. If it is in beverages, the bar operation needs attention. If it is distributed across everything, the problem is more likely to be loose purchasing records than any single operational issue.
What Changes After You Do This Consistently
The first stock take is always the most revealing, because there is no baseline to compare it against. By the third or fourth month, you are not just looking at the absolute variance. You are looking at whether the variance is improving, holding steady, or getting worse, and whether anything changed operationally in the months where it moved.
That trend is the information. It tells you whether the systems in the kitchen are getting tighter or whether there is something you have not addressed. No other reporting gives you that signal as clearly as a consistent monthly stock take.
It takes about ninety minutes to run ours. It is ninety minutes we do not skip. If you are not sure your operation has a food cost problem worth this kind of attention, these six signs are a faster first check.