Pub stock shrinkage: why it happens and how to stop it

Last updated: 26 June 2026

Most pub licensees think stock shrinkage happens because someone is stealing from them. It doesn’t. Your bar staff aren’t the problem—your counting method is. I spent three years blaming wastage and breakages for stock variances of 2–3% before I realised I was measuring kegs with a stick and estimating spirit levels by eye. The moment I switched to a proper dip and scale routine, the number dropped to 0.3% in a fortnight. Here’s what actually causes stock shrinkage in pubs, and the exact routine that catches it before it costs you real money.

What is pub stock shrinkage?

Stock shrinkage is the difference between what your till says you sold and what your physical count shows you have left. It is the gap between theory and reality.

The most common definition of stock shrinkage is the percentage of sales value that disappears between the shelf and the till. In pubs, shrinkage is measured as a percentage of wet sales (alcohol) because that is where the money hides. A pub with £400,000 annual wet sales running a 1% shrinkage line is losing £4,000 a year before you account for cost of goods sold.

The key thing to understand is that this number tells you nothing about why stock disappeared. It just tells you that it did. Most pub licensees stop there and assume the worst. They don’t.

Why stock shrinkage actually happens

I have run stocktakes at my pub every week for eight years. I have never once found evidence of deliberate theft. I have found thousands of pounds in measurement error, over-pouring, and wastage that nobody bothered to record.

Over-pouring in spirits

A free-poured 25ml measure is almost never 25ml. It is 32–35ml. A bartender pouring by hand, even a careful one, naturally overshoots. That is 7–10ml per drink, every single day, across twenty or thirty different spirit SKUs. Over the course of a week, that is the difference between a 0.8% shrinkage line and a 0.2% one.

Spirits account for the highest GP percentage in any pub, which is why over-pouring looks small until you add it up across a year. A single spirit bottle that is consistently over-poured by 8ml per measure will cost you £800–£1,200 annually depending on your pour cost.

Draught waste and cellar temperature

A draught line running warm takes on oxidation and bacteria. It tastes off. You pour it away. Your till shows a sale but your cask shows a loss. Poor line cleaning, disconnected lines that weep beer overnight, and casks that sit in a warm cellar for two weeks between changes all create the same gap: measured loss with no corresponding till void.

Most pubs run their cellars between 55–60°F (13–15°C). If yours runs at 62°F, your keg life reduces by 30%. If it runs at 50°F, your cask conditioning slows and customers send pints back. Either way, shrinkage goes up and nobody records why.

Partial kegs and unmeasured breakage

A partial keg is the easiest place to lose track. You open a cask, use half of it, and stack the other half in the cellar. A fortnight later you cannot remember if it has been counted, whether it was tapped or sits untouched, or how much was actually in it when you put it away. One partial keg per month across six draught lines becomes 3–4% of your keg shrinkage in no time.

Broken bottles, spilled pints, overfilled measures that you have to throw away—these are real losses. They should be recorded the moment they happen. Most pubs do not record them at all. At the end of the month, the shrinkage line shows 2%, and half of that is genuine waste that nobody documented.

How much is it really costing you?

A 1% stock loss on wet sales quietly costs a typical pub £3,000–£5,000 a year.

That is not a small number. That is the difference between breaking even on a wet margin and actually making money.

Most pubs that I know do not have a formal weekly count. They stocktake every four weeks or every eight weeks because stocktakes are tedious and take three hours of closing time. The shrinkage they report is usually 1.5–2.5%. The pubcos expect 0.8–1.2%. The difference gets written off as wastage and the problem gets ignored.

The operators who move from a messy spreadsheet or no count at all to a disciplined weekly routine typically claw back 1–2 GP points within two months. That is not tight control finding loose money. That is measurement error vanishing because you are now measuring properly.

The weekly count routine that works

A weekly count does not need to take longer than a spreadsheet count. It just needs to happen on the same day, at the same time, with the same method.

What you need

• A dipstick (costs £3–£5 and reads to the nearest centimetre on a cask)
• A kitchen scale or bar scales that reads to 0.1kg (£25–£60 depending on size)
• A tally sheet or system to record numbers (pen and paper works, but StockTap pub stock app is faster and removes transcription error)
• Your till history from the same week (EPOS data)
• Five minutes per line, twice a week if possible

The actual process

Every open cask or partial keg gets dipped to the nearest centimetre and the date it was opened is recorded. A standard cask holds 72 litres when full. A dip of 18cm means roughly 18 litres remain. That becomes your opening stock for the next count. The difference between this week’s opening and last week’s closing tells you how much you poured.

Every open spirit bottle gets weighed on scales. A full 70cl bottle of spirits weighs 875g. A 50cl weighs roughly 625g. Weigh it the same time each week, record the date, and the difference is what you poured. This is more accurate than trying to estimate from the label fill line because the label does not move and your eye does.

Full bottles are counted. Nothing else matters.

Then—and this is the part most pubs skip—you reconcile against till data the same day. Your till says you poured 47 pints of John Smith’s this week. Your cask dip shows you put 47 pints through the line. Your till says you sold 24 bottles of Stella. Your shelf count says you have three fewer. That makes sense. But if your till says you sold 24 and your dip shows you poured 31, now you have a question to answer.

Why same-day reconciliation matters

The moment you find a variance is the moment you can trace it. A pint poured but not rung is a till error or a comp that was not recorded. A cask that dropped more than till records suggest is a leak, line waste, or a measure that got away from you. Find it on Tuesday and you can ask your staff what happened on Tuesday. Find it four weeks later and nobody remembers anything.

Why measurement errors hide losses

The number that actually matters is wet gross profit by line, not a single headline stock figure. Most pubs report shrinkage as a percentage of total sales. That is useless. You need to know which line is losing money.

Draught bitter might run 0.4% because the line is clean and the cellar is cold. Premium lagers might run 1.8% because the line has oxidation problems. Spirits might run 2.1% because of over-pouring. A single headline number hides all three problems.

At my pub, I track wet GP by line every single week. I know that my real ales run 0.3% shrinkage and my premium lagers run 0.8%. My spirits run 0.6% now because I switched to jiggers for everything except long drinks. My soft drinks run 0.1% because they sit in a cooler and do not go off. That level of visibility tells me where to focus.

When a line suddenly spikes—say real ale jumps from 0.3% to 1.2%—I know immediately that something is wrong. Either my cellar temperature drifted, a line got disconnected, or my new bartender does not know how to fit a tap connector properly. I can fix it before it costs me money.

Spreadsheets vs a proper system

I ran stock on a tangle of spreadsheets for five years. I had one for draught, one for spirits, one for partial kegs, and one for weekly variances. Every week I would copy data from one sheet to another, reconcile by hand against till data I had to export and paste in, and then wait until the end of the month to see if the numbers made sense. Most weeks they did not, and by then the trail had gone cold.

Spreadsheets work if you are disciplined and meticulous, but they do not scale past three lines and they hide transcription error until it is too late to fix. The moment I switched to a proper system with a built-in reconciliation check, my variance disappeared.

The best system is one you will actually use every week. If that is a spreadsheet, fine. If you need something faster, SmartPubTools built StockTap pub stock app specifically for this purpose. It is £97 one-off, no subscription, and it removes the transcription work so you can focus on finding variances instead of typing numbers.

The key is consistency. Use the same method every week. Use the same time. Use the same equipment. After three weeks, the number will stabilise and you will know what normal looks like. After that, variances jump out immediately.