What causes pub stock discrepancy

Last updated: 26 June 2026

A 1% loss on wet sales costs you between £3,000 and £5,000 a year, and most pubs never spot it until their margin report lands like a brick on the desk. You’re not being robbed — you’re just not measuring properly. That’s the uncomfortable truth behind most pub stock discrepancies. I spent years running my cellar on a tangle of spreadsheets and partial guesses, losing track of kegs mid-rotation and spirit bottles that never made it to the till. When I finally sat down with a dipstick and a set of scales to do a proper count, the variance dropped from complete guesswork to a number I could actually trust within a fortnight. This article shows you exactly what a pub stock discrepancy is, why they happen, and how to catch them before they become invisible profit drains.

What Is a Pub Stock Discrepancy

A pub stock discrepancy is the difference between the stock value your EPOS system recorded as sold and the actual stock count you find in the cellar. If your till says you poured 50 pints of ale on Monday, but when you count the cask it’s lost 55 pints, you have a 5-pint discrepancy. Multiply that across draught lines, spirits, lager, cider, and bottled stock, and you’ve got a figure that’s either eating your margin or hiding a real operational problem.

Most pub managers think of stock discrepancy as theft. In practice, it’s rarely that simple. A discrepancy sits on a spectrum. At one end you have genuine shrinkage — evaporation, settling in kegs, natural wastage. In the middle you have measurement error — dipping a cask wrong, forgetting to weigh an open spirit bottle, not accounting for line purge after cleaning. At the other end you have genuine loss: over-pouring, forgotten wastage that never got logged, or actual stock going missing. Before you blame staff, you need to know which one you’re looking at.

Why Stock Discrepancies Actually Happen

I’ve run stock takes at my pub with three different systems over 15 years, and I’ve seen the same causes repeat every time. Here’s what actually drives the gap:

Draught Beer and Cask Management

Draught is the biggest culprit because there are so many places for loss to hide. A cask sits at 55°F (13°C) in most UK pubs, but if your cellar temperature creeps to 58°F (14°C) or above, you get faster settling and more natural evaporation. You pull a pint that’s 2mm too foamy because the pressure is off. You clean the lines on a Friday and don’t account for the beer that sits in the line overnight — that’s typically 1.5 to 2 pints of waste per line that most spreadsheets miss entirely. You rotate a cask and don’t mark it as ‘in stock’ until Tuesday, even though it’s been pouring since Monday lunchtime. A single cask in a busy pub can hide 5–8 pints of unaccounted variance.

Spirit Measures and Over-Pouring

This is where measurement error becomes invisible cost. A 25ml measure should pour exactly 25ml, but a free-poured spirit in a busy shift is often 32–35ml. You can’t spot this in a till variance because the till recorded a 25ml sale either way. The only way you catch it is to weigh every open spirit bottle at the same time each week and reconcile against your till count. A single spirit line losing 7ml per pour across 40 pours a day is 280ml unaccounted for — that’s nearly a full bottle a week of ghost loss.

Forgotten Wastage Logs

You break a bottle. Someone spills a pint. A keg is damaged and you have to pour it away. None of these should hit your stock count as ‘loss’ — they should hit your wastage log and come off cost of sales separately. But most spreadsheet-based pubs don’t have a reliable wastage log, so these items either get ignored or half-remembered three weeks later.

Till Reconciliation Lag

Your EPOS system might record that you sold 120 pints of lager on Tuesday, but those 120 pints weren’t all on the same cask. Maybe 80 came from cask A and 40 from cask B. If you don’t reconcile your till to your cellar on the same day, you end up dipping two casks days later and trying to remember which one was swapped when. By then, you’ve lost the data trail.

Measurement Error vs Real Loss

Here’s the operator insight that changed my whole approach to stock: most stock ‘theft’ is actually measurement error and forgotten wastage, not cash register crime. When I started using a proper dipstick and scales, I found I’d been dipping casks wrong for years. I was measuring from the top instead of the side, which changes the read by 2–3 pints. I wasn’t weighing open spirit bottles — I was just eyeballing how much was left. And I had no idea how much beer sat in the lines after cleaning.

Once you fix the measurement, you can actually see real loss. A pint that disappears between the till and the cask isn’t measurement error — it’s a training problem, a pouring issue, or an actual stock movement you didn’t log.

The way to tell the difference is to run a tight count for three weeks and watch the pattern. If the discrepancy is consistent and small (under 0.5%), it’s almost certainly measurement error or natural settling. If it’s random and erratic (3–5% some weeks, -1% others), you’ve got a control issue — different staff members counting wrong, or variance in how the count is recorded. If it’s consistently high and one-directional (always a loss, never a gain), you’ve got a real problem: over-pouring, or something going missing.

Building a Weekly Count Routine

The best way to catch a pub stock discrepancy before it compounds is to count your stock every week, same day, same time. This is not a monthly stocktake. This is a 30-minute line check that becomes part of your operating routine. Here’s what I do at my pub:

Monday Morning, 10am

• Dip every cask and partial keg on the bar. Use a proper dipstick, measure from the side, not the top. Record the dip depth and the cask reference. If a cask has been on for less than 24 hours, note the date it went on.
• Weigh every open spirit bottle. Use a set of scales, record the weight, and compare it to the weight from the same time last week. The till tells you how many 25ml pours it recorded — the weight tells you what actually came out of the bottle.
• Log any wastage from the previous week. Broken bottles, spoiled stock, line purge from cleaning, kegs sent back damaged. This is not loss — this is cost of sales that needs to be separated from your margin calculation.

Same Day, Before Lunch Service

• Pull your till report for the last seven days. Sort by line: how many pints of each draught were recorded as sold? How many 25ml measures of each spirit?
• Reconcile the till against the cask dips and spirit weights. If the till says you sold 200 pints of lager and your casks show a combined dip of 210 pints, you have a 10-pint discrepancy. Write it down. If it happens three weeks running, something’s wrong with measurement or pouring control.

This routine takes 30 minutes and gives you a number you can actually trust. After three weeks you’ll see a pattern. If most lines sit between +2 and -2 pints variance, you’re tight. If one line is consistently out by 8–10 pints, you’ve found a training issue or a temperature problem. If a spirit line is losing 15ml per dip consistently, you’re over-pouring.

At my pub, once I switched from a tangle of spreadsheets to this disciplined approach, my weekly variance stabilised within a fortnight. I was able to see that my cask ale line was losing 3–4 pints a week to line cleaning waste I wasn’t accounting for, and my vodka line was consistently short because I wasn’t using a measure on busy nights. Neither of those problems would have shown up in a monthly stocktake.

Using Technology to Track Stock

A spreadsheet can work if you’re disciplined, but most pubs aren’t. You miss a week, you forget to update the spirit weights, you don’t log wastage properly, and suddenly your data is useless. StockTap pub stock app was built specifically to solve this problem by automating the routine and making sure the same data gets recorded every week without gaps or guesswork.

The advantage of a purpose-built system over a spreadsheet is that it forces consistency. Every Monday at 10am, you get a reminder. You dip the cask, you enter the number, the app stores it alongside last week’s dip, calculates the variance, and flags anything outside your normal range immediately. You weigh the spirits, it tracks the trajectory. You enter your till data, it reconciles automatically.

Is an app safer than a spreadsheet for your records? Yes, for one simple reason: a spreadsheet lives on one person’s laptop. If that laptop dies, or that person leaves, the data goes with them. An app stores your data in one consistent place, and you can access it from any device. You can also see trends across months and spot seasonal patterns — like the fact that your stock variance gets worse in summer when the cellar temperature drifts. SmartPubTools also backs up automatically, so you’re not one crashed hard drive away from losing three months of records.

Red Flags and When to Investigate

Not every discrepancy needs investigation. A +1 or -1 pint variance on a cask is noise. But here’s what should trigger a deeper look:

• One line consistently out by more than 3% — this is a pouring control issue, a temperature issue, or a measurement problem specific to that line. Pull the staff rota and see if one person works that line. Watch them pour for a shift.
• A cask that’s consistently short before it should be empty — this is often a settling or temperature issue. Check your cellar temperature and cask pressure. Ask your brewer about the cask condition when it arrived.
• An open spirit bottle losing weight faster than the till explains — this is either over-pouring or free-pouring instead of using a measure. Retrain on measures immediately and increase the frequency of your checks to daily for a week.
• Wastage suddenly spiking without explanation — if your logged wastage jumps from £20 a week to £80, something’s changed. Either there’s genuine breakage you need to investigate (are you stacking kegs wrong? is your fridge thermostat going?), or stock is being poured away without being logged.
• Your overall stock variance jumping above 2% in a single week — this is usually a data entry error, a cask you forgot to reconcile, or a delivery you received but didn’t count in. Go back and check the data before you assume there’s a real problem.

The point is this: a weekly routine gives you visibility. You see trends instead of shocks. You catch the 5-pint loss in week one, not the 200-pint loss three months later.