Every bar has two pour costs (CMV, as pour cost is called in Brazil). One is what the drinks should cost, based on the recipe cards. The other is what the bar actually spent. Comparing theoretical vs actual pour cost is one of the most useful checks a bar can run, because the gap between the two is money that left the building without being charged for. This article explains both numbers, works through a month of example data, and lists the usual reasons they do not match.
The figures use drinks costed by the CostFlow engine for a fictitious bar with fictitious purchase prices, in Brazilian reais; the math is the same in any currency. The month itself is also fictitious.
Theoretical vs actual pour cost: two different questions
Theoretical pour cost answers: if every drink sold had been made exactly to the recipe card, what share of sales would have gone to ingredients? You get it from the card cost of each drink, multiplied by how many were sold, divided by sales.
Actual pour cost answers: what share of sales did the bar really consume in stock? You get it from inventory: opening stock, plus purchases, minus closing stock. That is the cost of everything that left the shelves, for any reason.
The first number comes from the recipe cards and the sales report. The second comes from the stockroom and the invoices. They are built from different sources, which is exactly why comparing them is useful.
Theoretical pour cost for one month
Take a month in which the bar sells three drinks. The cost per drink comes from the recipe cards; see how to calculate cocktail pour cost for how each one is built.
| Drink | Sold | Menu price | Sales | Theoretical unit cost | Theoretical cost |
|---|---|---|---|---|---|
| Gin & tonic | 300 | 28.00 | 8,400.00 | 6.66 | 1,998.00 |
| Negroni | 200 | 32.00 | 6,400.00 | 7.08 | 1,416.00 |
| Caipirinha | 250 | 16.00 | 4,000.00 | 2.23 | 557.50 |
| Total | 18,800.00 | 3,971.50 |
Theoretical pour cost = 3,971.50 ÷ 18,800.00 = 21.13%.
This is what the month should have cost if every drink had been made to spec, with the waste allowance already built into each card.
Actual pour cost for the same month
Now the stockroom. At cost value:
- Opening stock: 3,200.00
- Purchases during the month: 5,000.00
- Closing stock: 3,500.00
Consumed = 3,200.00 + 5,000.00 − 3,500.00 = 4,700.00.
Actual pour cost = 4,700.00 ÷ 18,800.00 = 25.00%.
The gap
The difference is 25.00% − 21.13% = 3.87 percentage points.
In money: 4,700.00 − 3,971.50 = R$ 728.50 in the month. That is stock the bar paid for and that does not correspond to any drink sold at its recipe cost.
A gap of zero is unlikely in any real bar. The question is how big the gap is, whether it is stable from month to month, and where it comes from.
Why the numbers don't match
The usual causes, roughly in the order worth checking:
- Pours above the recipe. If the card says 50 ml and the bar pours 55 ml, every drink uses more than the theoretical cost assumes. Small extra pours across hundreds of drinks add up.
- Unrecorded waste. Breakage, spills, remakes and prep leftovers that go beyond the allowance on the cards. More on this in bar waste and pour cost.
- Comps. Drinks given away consume stock but generate no sales. If they are not recorded, they look like a cost problem.
- Sales not entered. A drink made and paid for but not rung up in the system leaves stock without leaving sales. The theoretical side never sees it.
- The card differs from the bar. The office costs one recipe, the bar makes another: a different brand, a bigger pour, an extra ingredient. The fix is one spec for both; see the cocktail recipe card template.
- Outdated purchase prices. If the cards still use last quarter's prices and suppliers raised them, the theoretical cost is too low, and the gap is partly on paper.
- Counting errors. A miscounted bottle in the opening or closing stock moves the actual cost directly. A wrong count can create a gap, or hide one.
Not every cause is a loss in the stockroom. Some are errors in the numbers themselves. That is why the next step is to check the count before drawing conclusions.
How to close the count
A gap is only meaningful if the count behind it is solid. Three habits make it so:
- Count at the same day and time. Count opening and closing stock at the same point in the week and the same hour, ideally before service. A count taken in the middle of a shift mixes what is on the shelf with what is in the glass.
- Check purchases against the invoices. Use the delivered quantities and the prices on the invoices, not the order or the price list. Deliveries that arrived but were not entered, or were entered twice, distort the result.
- Compare by product before the total. Do the theoretical vs actual comparison bottle by bottle. How much gin should the month have used, given the drinks sold? How much did it use? A total gap of 3.87 points says something is off. A product-level comparison says where.
Read the gap product by product
Working bottle by bottle changes the conversation. If the gap concentrates in one spirit, look at the drinks that use it: pours, specs, comps. If it is spread evenly across everything, look at the count, the invoices and the prices first. If it appears in one month and not in others, look at what was different that month.
The goal is not to blame anyone. It is to make the theoretical number and the actual number describe the same bar.
Know where your gap comes from
Ambar is preparing a pour cost diagnostic for bars, built with the CostFlow engine from your recipe specs and your purchase prices. It opens soon, at the end of November. Leave your e-mail on the /motor page to join the list and hear when it opens.