Bar inventory variance: why it matters

Last updated: 26 June 2026

Most pub licensees have no idea what their actual stock loss is — they just know the number feels wrong. A 1% stock loss on wet sales quietly costs a typical pub £3,000–£5,000 a year, and the majority of licensees don’t spot it until it’s too late. Bar inventory variance isn’t a problem that fixes itself. It’s the gap between what your till says walked out the door and what actually left your cellar, and understanding it is the difference between running a pub that makes money and running one that bleeds it.

If you’re running on a tangle of spreadsheets, guessing at partial kegs, or hoping your brewery stocktaker catches the holes, you’re not in control of your numbers — your numbers are in control of you. This article will show you exactly why bar inventory variance happens, how to measure it properly, and what to do about it before it costs you another year’s GP.

What is bar inventory variance?

Bar inventory variance is the difference between what your till says you sold and what your physical stock count shows you actually had available. In other words: your EPOS records a £100 shift, but when you count the bottles, kegs, and casks, you’re £8 short. That £8 is variance — and it’s money that left your pub without going through the till or being written off as wastage.

Every pub has some variance. The question is whether it’s 0.5% (acceptable), 2% (you’ve got a problem), or 5% (your cellar is leaking profit). Most licensees don’t even know their variance number because they’re not counting regularly enough to spot the pattern. They do a full stocktake every three months or once a year, which is why they never see the problem building.

Variance isn’t the same as stock loss — it’s the invisible bit of stock loss. You can see breakage. You can spot a missing bottle. But variance is the creep: a fraction of a millilitre here, a slightly generous pour there, a temperature swing in the cellar. Individually invisible. Collectively, it’s thousands.

Why bar inventory variance matters to your bottom line

Here’s what most new licensees don’t understand: your EPOS is not your friend when it comes to profit. Your EPOS tells you what sold. It does not tell you whether you made money. Bar inventory variance is the proof.

Let’s say your wet sales are £50,000 a month. A 1% variance means £500 disappeared without a till record or a written-off wastage note. At the end of a year, that’s £6,000 in unaccounted stock — roughly equivalent to losing a full point of gross profit. For a pub on a 60% wet GP margin, that’s the difference between 60% and 59%, which feels small until you realise it’s the difference between a pub that works and one that doesn’t.

Most pubs that move from a messy spreadsheet to a disciplined weekly count claw back 1–2 GP points within a couple of months. That’s not because they suddenly got better at pouring. It’s because they finally saw the problem and could address it. Visibility drives behaviour. Once your staff know you’re counting spirits by weight and dipping every cask, over-pouring stops fast.

At my own pub I was running stock on a tangle of spreadsheets and still losing track of partial kegs and spirit measures. The variance was sitting at around 2.5% on wet sales — invisible to me because I wasn’t counting weekly. I built a simple count routine around a dipstick and a set of scales, and the weekly variance went from guesswork to a number I could trust within a fortnight. Once I could see it, I could fix it. Within eight weeks, I’d clawed back nearly a point of GP just by getting the visibility right.

Common causes of bar inventory variance

Before you start blaming staff theft, understand this: the vast majority of variance is not theft. It’s three other things.

Over-pouring on spirits

A free-poured 25ml spirit measure is almost never 25ml. It’s 32–35ml. Your staff aren’t doing it maliciously — they’re doing it because pouring by eye is fast and they’ve never been trained on a scale. One free-pour per spirit per shift across a week adds up to two or three full bottles of unrecorded stock loss. Multiply that by your spirit range (vodka, gin, rum, whisky, brandy, liqueurs) and you’re looking at a permanent 0.8–1.2% variance just from pouring. This is the single biggest source of variance in most pubs.

Cellar temperature and line condition

A cask of bitter sits in a warm cellar and loses gas and liquid due to poor condition. You count it as a full unit when actually 10–15 litres has already evaporated or been lost to foam waste during service. A dirty or incorrectly fitted draft line spits foam instead of beer — that’s stock loss disguised as wastage. Temperature swings in summer cause similar issues. A 1-degree rise in cellar temperature can shift cask losses by 5–10% across a full inventory.

Measurement error and forgotten wastage

You pour a spirit sample for a customer and forget to ring it. You drop a bottle and don’t write it off. You adjust a pour because it looked wrong and don’t log it. Your cellar manager tops up a partial keg and doesn’t note it. These are not theft — they’re process gaps. Most stock ‘theft’ is actually measurement error and forgotten wastage. Fix the process and the variance drops.

Supplier short-measures and delivery errors

Less common but real: a cask arrives short, or a case of spirits is one bottle light. You receive it without checking the invoice properly. Six months later you’re wondering why your variance is high. Always count inbound stock against the docket the same day.

How to measure bar inventory variance properly

You need three inputs: what you had at the start of the period, what you received, what you sold, and what you physically have now.

The formula is simple:

Variance % = (Theoretical Stock – Actual Stock) ÷ Theoretical Stock × 100

Where:

• Theoretical Stock = Opening stock + Deliveries – Till sales – Recorded wastage
• Actual Stock = What you physically count

If your theoretical stock is 240 bottles of spirits and you count 235, your variance is 2%. If your theoretical stock is 150 casks of real ale at the week start and you’re down to 146 after delivery and till activity, that’s 2.7% variance on draught.

The number that actually matters is wet GP by line, not a single headline stock figure. Spirits might be running 1.5% variance, but draught might be 3% because of cellar temperature. Lager might be tight, but cider might be loose. You need to count by category and trend the numbers weekly, not monthly or yearly.

This is where most licensees go wrong: they do a full stocktake once a quarter and get a headline number that tells them nothing useful. A 2% variance over 12 weeks could mean 0.5% every week (normal), or it could mean you’re losing 4% some weeks and gaining 1% others (chaos). Only weekly counting shows you the pattern.

How to fix variance fast

Once you know your variance number, you need to attack the three biggest sources: over-pouring, cellar condition, and process gaps.

Weighing spirits as your control

Buy a set of digital scales (£15–30). Every time you take a stock count, weigh open spirit bottles. A full 70cl bottle of standard spirit weighs 790g. A half-full bottle weighs roughly 395g. You can tell at a glance whether your bartenders are pouring 25ml or 35ml — and once they know you’re checking by weight, the pouring gets tighter immediately. You’ll catch over-pouring within a week.

Dipping every cask and partial keg the same day

A dipstick costs £8. It takes 30 seconds per cask. You get a direct measure of liquid depth, which converts to litres. Do this every week on the same day, before service. It removes guesswork and catches temperature-driven loss and line-condition issues instantly. If a cask is dropping 5 litres a week without matching till sales, you’ve got a line problem or a leak.

Reconciling against till data the same day

Count your stock on Tuesday morning. Pull your till Z-read from Monday evening. Compare them the same day, while the numbers are fresh. Don’t wait a week. Don’t wait until month-end. Same-day reconciliation means you catch errors early and can go back to staff who were on duty to understand what happened. A £20 discrepancy on Tuesday is fixable. A £200 discrepancy you only spot three weeks later is not.

Use SmartPubTools to track this easily. The StockTap pub stock app lets you log your counts, compare them against till data automatically, and trend your variance over weeks so you can see patterns emerging. No subscription. No per-month fees. One-off tool built by someone who actually runs a pub.

Training your team on portion control

Your staff don’t over-pour because they’re trying to lose you money. They over-pour because they’ve never been taught the difference between a 25ml, 35ml, or 50ml measure. Show them. Make it visual. Use a scale during a team talk: show them what 25ml vodka weighs, what 35ml looks like. Tell them you’re counting by weight weekly. Suddenly, consistency improves.

Building a weekly line check routine

If you want to stay on top of variance, it has to be routine. Here’s what works:

Tuesday morning, 30 minutes

• Pull your till Z-read from Monday evening (sales by category)
• Count all open spirit bottles and weigh them
• Dip every cask, partial keg, and container of cider or wine
• Note any breakage or written-off wastage from the previous week
• Log it all in a spreadsheet or app (paper counts get lost)
• Calculate variance: (Theoretical – Actual) ÷ Theoretical × 100
• Spot-check any line that’s high variance — is the cellar temperature right? Is the line clean?

30 minutes a week beats a three-hour panic stocktake every three months. You catch problems early. You have a running variance trend you can actually act on. And you signal to staff that stock control matters.

Most licensees say they don’t have time for this. What they mean is they haven’t realised that not doing it costs them 1–2 GP points. If you’re spending £50,000 on wet stock a year and losing 1% of it to variance, you’re paying 3 hours a week in lost profit to avoid 30 minutes of counting. The maths are broken.

Monthly trend review

At month-end, look back at the four weekly variance numbers. Are they trending down (good)? Are they stable (acceptable)? Are they drifting up (you’ve got a new problem)? If spirits are running 1%, draught 2.5%, and cider 0.8%, you know where to focus: draught. Check the line condition. Check the cellar temperature. That’s where your money is leaking.

Quarterly full stocktake

Still do a full stocktake once a quarter. But now it’s a sanity check on your weekly counts, not a surprise audit. If your weekly variance is steady at 1.5% and your quarterly stocktake comes in at 1.4%, your system is working. If your weekly averages say 1% but your quarterly shows 3%, something went wrong — either in your weekly counting or in a period you didn’t spot.