Pour cost: the number behind every drinks price

Pour cost is the share of a drink's price that goes on what is in the glass. A gin and tonic that sells for 34 and uses 4.60 of gin, tonic, lime and ice has a pour cost of about 14 per cent once the tax is taken off the price. It is the bar's version of food cost percentage, and like that figure it is easy to calculate and easy to misread.

The usual mistake is to look at one number for the whole bar. A single figure for a week of trade blends cheap beers with expensive wines, cocktails with soft drinks, and careful pours with heavy ones. It can sit at a comfortable level while three or four products quietly lose money, and it can rise without anything being wrong at all.

This guide works the sum out at three levels: for one drink, for each category, and for the bar as a whole. It shows how to cost a cocktail properly, how a few millilitres of over pour add up, how to price from the percentage without letting it run the menu, and which products a healthy headline figure tends to hide. Every figure in it is an example, not a benchmark.

The figure answers one question: of every unit of currency a guest spends on a drink, how much goes on the liquid, the mixer, the garnish and the ice? It leaves out the bartender's time, the glass, the rent and the music. It is a ratio, so it is good at showing change and comparing like with like, and weak at telling you whether a drink is worth selling.

You will also see it called beverage cost percentage. The two names usually mean the same sum. Some businesses use beverage cost percentage for the drinks line of the profit and loss account, soft drinks and coffee included, and keep the shorter name for the bar's alcohol. Pick a definition, write it down and keep it, because a figure that changes its meaning from month to month cannot be compared with itself.

The kitchen has the same idea under another name, and our guide to food cost percentage walks through it with the same care. What is different at the bar is that drinks are poured rather than plated. A measure is small, it is often poured by eye, and a few millilitres either way are invisible to everyone except the stock count.

The starting point is one drink, made to its specification. Write down what goes in the glass, cost each part, add them up and divide by the price.

Here is an illustrative house gin and tonic, with example prices.

The drink costs 4.60 to make. It sells for 34, and that price includes tax at 5 per cent, so the price before tax is 34 ÷ 1.05, or 32.38. The result is 4.60 ÷ 32.38 × 100, which is 14.2 per cent.

Two things in that sum are easy to leave out, and both make the result look better than it is. The first is the mixer and the garnish, which are small on one drink and large across a year. The second is the tax inside the price, which the next section deals with.

The price a guest pays usually includes VAT or sales tax, and sometimes a service charge or a local fee. None of that is yours to keep, so none of it belongs in the bottom half of the sum. Divide by the price before tax and before any charge added on top.

In the example, dividing 4.60 by the full 34 gives 13.5 per cent rather than 14.2. The difference looks small on one drink. Across a bar, it means measuring yourself against a target you are not really meeting, and it hides a point of cost that is there every week. If your prices include several taxes or charges, remove them all, and work from the same net figure your accounts use for drinks revenue.

A cocktail has more parts than a spirit and mixer, some of them made in house, and that is where most costings go wrong. The method is the same as for any recipe: write the specification as the drink is really made, cost each part per millilitre or per piece, and add them up. Our guide to recipe costing covers the kitchen version step by step. To cost a cocktail, the bar adds a few habits of its own.

Here is an illustrative whisky sour, again with example prices.

The drink costs 8.77. At 52 including 5 per cent tax, the price before tax is 49.52, and its pour cost is 17.7 per cent. Notice where the money is. The bourbon is four fifths of the cost, and the next largest item is the egg, which most quick costings leave out.

Once the drinks are costed this way, keep the costs on the spec cards. A spec that lives only in a senior bartender's head cannot be costed, checked or taught, and it changes every time someone new learns it.

Different kinds of drink have different natural costs. Spirits with a mixer, cocktails, wine by the glass, wine by the bottle, draught beer, bottled beer and soft drinks are all built from products with different prices and different margins. A bar that sells mostly beer and a bar that sells mostly wine can both be well run and still show very different figures.

That is why the figure should be calculated by category, not only in total. The total is a blend, and a blend moves when the mix moves. Consider two illustrative weeks in which every category costs exactly the same:

Nothing went wrong in week two. Every drink was poured and priced exactly as before, and the headline still rose by nearly two points. The opposite happens too: a busy week of beer can pull the total down while the cocktail list leaks money underneath it. Read the categories first, and the total only as the sum of them.

Per-drink costs give you one version of the bar's figure, and the stock count gives you the other. Comparing the two is how a bar finds out where its money goes.

Theoretical is what the bar would have spent if every drink had been made exactly to its spec. Actual is what it did spend, measured from the count. The gap between them is the variance, and it holds everything the specs leave out: over pour, spillage, breakage, comps and staff drinks that were never recorded, drinks that were made and not rung through, and theft.

The actual figure needs a regular count. Our guide to bar inventory management sets out a weekly one that fits between last orders and lock up. The kitchen has the same pair of figures, explained in actual vs theoretical food cost.

A small gap is normal, because no bar pours to the millilitre. What matters is its size against your own history, and where it sits. A gap spread thinly across the whole bar is a different problem from one concentrated in a single product or on a single shift.

When a bar's total looks fine and its profit does not, the cause is rarely the whole bar. It is usually a short list of products whose cost is wrong in ways the total smooths over. Three come up again and again.

The spirit on the speed rail is poured more often than anything else in the building, usually by eye and often in a rush. A few millilitres over on each drink is invisible on the night and large over a year, precisely because the volume is so high. It hides in the total because the house spirit is cheap per bottle, so its excess looks small against the bar's sales.

An open bottle starts to lose its value the moment it is opened. Generous glasses, the last glass poured away, bottles opened for a taste and never sold: each is a cost that no specification includes. Wine by the glass often carries a respectable theoretical figure and a poor actual one, and only a count by category shows the difference.

Many bars price a premium spirit as the house drink plus a fixed amount. That works when the premium bottle costs a little more, and fails when it costs three or four times as much. The drink sells well, because it looks like good value, and every one carries a cost far above the rest of the list. In the total it is a minor line. In its category it is often the biggest single leak.

Look for your own versions of these three before anything else. They are not the only causes of a gap, but they are the ones a comfortable number hides most often.

Over pour is the gap between the measure in the spec and the measure in the glass. It rarely looks dramatic, which is exactly why it is worth putting into numbers. Go back to the house gin and tonic, where a 25 millilitre measure costs 3.00 and a 30 millilitre pour costs 3.60.

The fix is less about discipline than about tools and habits. Measured pours, with jiggers or measured pourers, make the spec the default. Free pouring suits some bars and their style, and it can be accurate, but it needs testing. A regular pour test with water, poured into a measuring cylinder, shows each bartender where their hand sits, and the bar inventory guide describes a simple version of it.

The percentage is also a pricing tool, and the arithmetic runs backwards from the formula. Choose the figure you want for a drink, divide the drink's cost by it to get the price before tax, then add the tax.

For the whisky sour, at an illustrative target of 20 per cent, that is 8.77 ÷ 0.20, or 43.85 before tax, which becomes 46.04 with 5 per cent tax added. You would round that to a price that reads well on the menu, such as 46 or 48. Our guide to menu pricing sets this method beside five others and explains when each one works.

The trap is letting the percentage run the whole list. It is a ratio, and ratios favour cheap products. Take an illustrative premium whisky: 50 millilitres from a bottle costing 420 costs 30.00. At 90 including tax, its price before tax is 85.71 and its cost is 35 per cent of the price. That looks expensive beside the gin and tonic at 14.2 per cent. But the whisky leaves 55.71 in cash after the cost of the pour, and the gin and tonic leaves 27.78. A guest who trades up to the whisky is worth twice as much to the bar, at more than twice the percentage.

So use the percentage to set a floor and to spot drinks that have drifted, and use the cash each drink leaves to decide what to put in front of guests. Our explainer on gross profit on a dish covers the same idea for food, and it applies to a glass exactly as it does to a plate.

A drink's cost is only as current as the prices behind it. When a supplier puts up the price of the house gin, every drink that uses it costs more, and the specs on file are suddenly wrong. If the 700 millilitre bottle in the example goes from 84 to 91, each millilitre costs 0.13, the measure costs 3.25, and the gin and tonic costs 4.85 to make. It moves from 14.2 to 15.0 per cent without anyone pouring differently.

Re-cost from the latest invoice whenever a price changes on one of your top sellers, and review every spec on a regular cycle, quarterly for most bars. Keep the date of the last costing on each spec card, so anyone reading it knows how current it is.

Bottled products are easy to cost, because the price of the bottle is the cost of what is in it. Wine by the glass and draught beer are harder. What you buy and what you sell are measured differently, and some of what you buy is never sold.

For wine by the glass, divide the bottle price by the number of glasses you actually sell from a bottle, not the number it could hold. A 750 millilitre bottle holds five 150 millilitre glasses. If, on average, part of the last glass is poured away or the bottle is opened for a taste, the real yield is lower, and each glass sold carries more cost. Record what is poured away for a few weeks, then cost from the yield you really get.

For draught, cost from the real yield of a keg after foam, line cleaning and the first pulls of a new keg. A keg costed as if every drop reaches a guest will always look better on paper than in the count. Line cleaning is a regular, legitimate loss, so record it and cost it as one rather than letting it sit in the variance.

The figure earns its keep when it is read the same way every week. A routine that takes less than an hour after the weekly count looks like this:

Keep the results in one place, week after week. A single week means little on its own. The line it draws across a quarter shows whether the bar is getting better.

Each of these makes the figure friendlier and less useful. The aim is not a low number. It is a true one.

The calculation needs two halves: what the drinks cost, and what the bar sold. servearound covers the second half, and the records that explain a gap. The first half stays with your specs, invoices and count.

Every drink rung through the point of sale lands in the menu report in Analytics, which shows units sold, revenue and share of sales for each drink and each category over any range of dates. That is the sales column for the theoretical cost and for the comparison with your count. The payments report breaks voids and comps down by member of staff, which is where unrecorded giveaways start to show. When the bar runs out of something, 86 the drink on the till and it shows as sold out on every table's menu at once, so nobody keeps selling a cocktail the bar cannot make.

servearound does not track stock or hold recipe costs today, so bottle prices and counts stay on your sheet. Inventory tracking, for the stock behind the menu, is coming. If you would like to see the sales side for yourself, get in touch.

There is no single good figure. A cocktail bar, a pub and a restaurant with a serious wine list run different costs for good reasons, and a target borrowed from another kind of bar is rarely meaningful. Measure your own, by category, for several weeks. Then work on two things: keeping each category steady, and narrowing the gap between actual and theoretical.

Usually, yes: both divide the cost of drinks by drinks sales. Some businesses use beverage cost percentage for every drink on the menu, coffee and soft drinks included, and the shorter name for the bar's alcohol alone. Either works, as long as you choose one definition and keep it.

No. Divide by the price before VAT, sales tax and any service charge. Including them makes the figure look lower than it really is.

Cost the batch, then divide by what it makes. Add up everything that goes into the syrup, measure the finished volume, and work out a cost per millilitre. Re-cost it when the price of the main ingredient changes.

They count in the actual figure, because the stock was used. Record them every time and they move out of the unexplained variance into their own line, where they can be managed as a policy rather than suspected as a loss.

That depends partly on where you trade. In the UK, gin, rum, vodka and whisky sold for drinking on the premises must be served in 25 or 35 millilitre measures, or multiples of them, using the same measure across the premises, with a notice telling guests which one. Mixtures of three or more liquids are exempt. The rule is in the Weights and Measures (Intoxicating Liquor) Order 1988. Elsewhere, check the rules where you trade. Wherever you are, write the measure into every spec, because every costing starts from it.

Not always, but it is less consistent, and consistency is what keeps the actual figure close to the theoretical one. If your bar free pours, test every bartender regularly with a timed pour into a measuring cylinder, and watch the variance on the house spirits.

Not on its own. A drink with a low cost and a low price can leave less cash than a premium drink with a high cost. Use the percentage to catch drinks that have drifted and to set a floor, and look at the cash each drink leaves when you decide what to promote.

Pour cost is what goes in the glass as a share of the price before tax. Work it out for each drink from a written spec, including mixers, garnishes, ice and anything made in house. Read it by category, not only in total, because the total moves with the mix. Compare theoretical with actual every week, and look first at the house pour, wine by the glass and premium drinks sold on a flat surcharge.

Then use it for what it is good at: catching drift and setting a floor for prices. Decide what to sell on the cash each drink leaves, and keep the costings current every time a supplier changes a price.