What it means
A bar sells a drink for 10 currency units and its measured recipe ingredients cost 2 units, so the theoretical pour cost is 20%; this is a cost percentage, not a gross profit percentage. For all drinks over a period, use beverage cost of goods sold divided by beverage sales, where opening inventory plus purchases minus closing inventory is the basic cost-of-goods calculation, adjusted for transfers and other movements where relevant.
A fictional bar begins with 10,000 units of beverage stock, buys 2,000 and ends with 9,000, so its simple beverage consumption is 3,000 units, and if beverage sales are 15,000 the actual pour-cost ratio is 20%. That arithmetic does not identify each cause of consumption, since drinks made for staff training, breakage, promotions and serving mistakes can affect actual stock; document legitimate non-sale uses rather than calling every difference theft.
Theoretical cost comes from recipes and sales mix: multiply each drink's planned ingredient quantities by current unit costs, then aggregate them over actual drinks sold, which estimates the cost if recipes and records are followed. Actual cost uses stock movements and can exceed theoretical cost if staff pour more than the recipe, ingredients spoil or counts are wrong, so a variance deserves investigation, not an automatic accusation.
A high pour cost means more of the selling price is consumed by ingredients under the chosen method, but a low one is not automatically better, because price, drink quality and customer value still matter. Gross beverage margin before other costs is approximately 100% minus pour-cost percentage when revenue and ingredient cost use consistent bases; it is not net profit, since labour, rent, service and card fees remain.
Beverage categories differ, so a wine list, beer taps and cocktail menu can have different costs and prices, and category or product results avoid a blended figure hiding the problem. Selling more premium drinks can change the overall percentage even if every recipe stays the same, a sales-mix effect, so compare like periods or explain product-mix shifts.
Inventory counts must be credible, and partial bottles are hard to measure by sight, so use a consistent count method covering unopened and opened stock and investigate material discrepancies. Purchases during a week do not equal ingredients consumed that week, because a large restocking order may stay on shelves, which is why the period formula uses beginning and ending stock, and transfers between outlets should be recorded on both sides with consistent valuation so stock moved to another bar does not look like unexplained consumption.
Taxes and service charges can change the sales denominator, so use a defined net beverage-sales basis rather than mixing tax-inclusive receipts with pre-tax ingredient costs, and label the measure on reports. Recipe prices change with supplier costs, so a drink that met a target last month may now have a higher theoretical cost; update purchase prices and recipe yields before changing menu prices.
Complimentary drinks should have an authorised record because they use inventory without generating beverage sales and can push the actual ratio up, spillage and spoilage need distinct codes because root causes lead to different fixes such as storage, portioning or procurement, and a cocktail's garnish and mixer can be material costs, so include the ingredients within the agreed cost scope. Pour cost is a management indicator, not a universal target, as menu concept, price point, promotions and product mix affect a sensible range, so avoid lifting a vendor's benchmark without local evidence.
Use the measure to ask what happened to cost, price and stock: pair recipe estimates with actual counts and sales, then test a specific explanation before acting.
In practice
Real-world examples.
Example
A drink sold for 10 units with 2 units of recipe ingredients has a 20% theoretical pour cost. The bar shows this figure beside each recipe so bartenders see the target. When the supplier raises the spirit price, the recipe cost is updated and the figure recalculated before the menu price is changed.
Example
A bar calculates period cost from opening stock, purchases and closing stock: 10,000 + 2,000 - 9,000 = 3,000 units. Dividing by 15,000 of net beverage sales gives 20%. The manager repeats the count with a second person to confirm the partial bottles.
Example
A manager records complimentary drinks separately before investigating actual-to-theoretical variance. The authorised comps log explains part of the gap and breakage codes explain more. Only the remaining unexplained difference prompts a closer look at pouring practice.
Formula
Calculation
Recipe pour cost % = recipe ingredient cost / drink selling price x 100. Period pour cost % = beverage cost of goods sold / defined net beverage sales x 100; reconcile stock transfers and adjustments.
Worked example: a drink sells for 10 units and its recipe costs 2 units, so 2 / 10 x 100 = 20%. For the month, opening stock of 10,000 plus purchases of 2,000 minus closing stock of 9,000 gives cost of goods sold of 3,000, and 3,000 / 15,000 x 100 = 20% on net beverage sales of 15,000. Suppose recipes applied to the drinks actually sold give a theoretical cost of 2,700, which is 18% of sales. The variance is 3,000 - 2,700 = 300 units, or 2 percentage points of sales. If 120 units are logged as authorised complimentary drinks and 60 as breakage, 300 - 120 - 60 = 120 units remain unexplained, which is 120 / 15,000 = 0.8% of sales and the proper focus of further checks.Case study
Seen in the real world.
In this entirely fictional case, Willow Bar sells a cocktail for 10 units and estimates 2 units of ingredients, a 20% recipe pour cost. Monthly counts show a higher actual cost ratio. The manager checks spillage, complimentary drinks, sales mix and count errors before changing staff procedures. The example makes no claim about a universal target.
Over the month, net beverage sales are 15,000 and the recipes applied to the drinks sold give a theoretical cost of 2,700, against an actual cost of 3,000 from the stock counts. The authorised complimentary log accounts for 120 units and recorded breakage for 60, leaving 120 units unexplained. The manager treats that 0.8% of sales as a prompt to re-measure pours and recount the cellar, not as proof of theft.
Watch out
Common mistakes.
- Using purchases alone as consumption for a period.
- Calling all actual-to-theoretical variance theft.
- Treating pour cost as the same as final profit.
Questions
People also ask.
What is the difference between actual and theoretical?
Theoretical uses planned recipes and sales; actual uses measured inventory consumption and sales.
Does a low pour cost prove a healthy bar?
No. Labour, rent, demand and drink quality also matter.
Why did it rise?
Ingredient prices, sales mix, waste, portions, promotions or counting errors can each affect it.
From the founder's library

Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%