Cellar management training for pub licensees

Last updated: 26 June 2026

Most pub licensees never learn cellar management properly — and it costs them £3,000 to £5,000 a year in undetected stock loss. I ran my own pub on a tangle of spreadsheets for three years before I realised the variance in my count was actually just poor discipline masquerading as mystery shrinkage. Once I built a simple weekly routine around a dipstick, scales, and till reconciliation, the guesswork stopped. The number actually matters: wet GP by line, not a single headline stock figure. This article covers the training you actually need — the kind a working licensee would teach you, not a corporate manual.

Why most pubs fail at cellar management

Cellar management training gets delivered by pubcos as a box-ticking exercise. You get a laminated sheet, a brief conversation about stock rotation, and then you’re left to figure out the rest. The problem is nobody teaches you why accuracy matters, or what to actually measure. Most licensees end up with a stock variance that swings wildly — 2% one week, minus 1% the next — and then they assume it’s just the nature of the business.

It isn’t. Wild variance means you’re not counting properly.

I spent three years running stock on a spreadsheet where I’d note down what the brewery stocktaker told me, cross-reference it against some handwritten notes, and call it done. My variance was all over the place. I blamed it on staff, on evaporation, on the weather. Then I realised the actual problem: I wasn’t weighing anything, I wasn’t dipping properly, and I was reconciling against till data from three days later. I was measuring noise, not stock loss.

The real cost of bad cellar management isn’t just the stock you lose. It’s the decision paralysis that follows. You don’t know whether to trust your numbers, so you over-order as a buffer, which ties up cash. You can’t spot patterns because the data is garbage. You can’t train staff on portion control because you don’t actually know what’s being poured. And every time a brewer or pubco stocktaker comes in and finds a discrepancy, you’re on the back foot.

The weekly count routine that works

The most effective way to control cellar stock is to dip every cask and partial keg, weigh open spirit bottles, and reconcile against till data the same day. That’s it. No spreadsheets with 47 columns. No colour-coded tabs. A routine you can run in 20 minutes, every week, on the same day and time.

Here’s what I do at my pub:

• Tuesday morning, 9am sharp. Same time every week. Your staff know when it’s happening. Consistency is what makes the numbers trustworthy.
• Dip every cask and partial keg in the cellar. Write down the dip in millimetres. A standard UK cask is 288 litres (63.8 gallons for you US operators). A dip of 50mm is roughly 5 litres out. Keep a chart on the cellar wall — you’ll start to see patterns within a fortnight.
• Weigh every open spirit bottle. A full 70cl bottle of whisky is roughly 770g. If it’s 650g, you’ve poured 15 measures (at nominal 25ml per measure). If the till says 18 measures, you have a problem — likely over-pouring, not theft.
• Count closed bottles and beer stock. Fast job. Just count.
• Pull yesterday’s till data. Same day. Not three days later. Cross-reference the bottles you’ve counted against what was rung through. The variance should be tight — usually within 1–2 measures on spirits, or one cask on draught.

When I switched from a monthly count to this weekly routine, the variance went from guesswork to a number I could trust within two weeks. Now I can tell you whether the issue is over-pouring, temperature, bad line cleaning, or a till error — and I can spot it before it costs me money.

Use StockTap pub stock app to log this data instead of a clipboard. You’ll have a clear record, you can spot trends over months, and you can see which staff member is dipping, when, and what the result was. The data quality improves just by logging it properly.

Temperature, line cleaning and wastage

Draught beer loss hides in three places: cellar temperature, line cleaning waste, and bad practice. Most licensees never measure any of it.

Cellar temperature

Beer is supposed to be stored at 12–13°C. If your cellar runs at 15°C or above, you’re accelerating oxidation and shortening shelf life. Every degree above 13°C reduces the life of the keg by roughly a week. Hot cellars also mean more pressure variance, which means inconsistent pours and more foam waste. I’ve seen pubs lose half a cask a week to excess foam because they never checked the thermometer. The cost is buried in “normal wastage” on the variance sheet.

Buy a cheap digital thermometer (£5) and log it weekly. If it trends high, call a refrigeration engineer. If it’s erratic (swinging from 12°C to 16°C), you have an airflow problem.

Line cleaning waste

If your lines aren’t cleaned weekly with a proper cleaner and flushed correctly, you’re losing beer to dead legs and residue. Most pubs do a rinse and call it done. A proper clean takes 15 minutes per line and costs about £1 in cleaner. The payback is maybe a cask a month in reduced waste. If you have four draught lines, that’s four casks a year — roughly £200 at cost.

This is training that should happen hands-on. Get your person who’s doing the cleaning to watch a YouTube video on proper keg line cleaning (search “beer line cleaning procedure”), then get them to do it under supervision once. After that, spot-check monthly.

Wastage tracking

Every time a staff member pours a pint and it’s flat, or a keg kicks and you have to purge the line, or a bottle gets dropped — that should go in a wastage book. Not as punishment. As data. Most pubs never track this, so they don’t know whether it’s a real problem or normal. I log it in the same app where I log stock counts. After three months, I can see patterns — like “every Tuesday afternoon the flat pint count spikes” (which means the line pressure’s dropping at certain times) or “new barstaff has double the spillage rate” (which means they need better training on pouring technique).

Spirits, over-pouring and the scales

Spirits hide losses in over-pouring more than any other line. A free-poured 25ml measure is often 32–35ml — that’s a 40% loss of margin on every pour. If you’re running two bars and each barstaff member free-pours 30 spirits a night, that’s 60 spirits at 7ml over spec. Seven millilitres × 60 = 420ml a night. That’s one bottle every four days, just from being generous with the pourer.

Spirits require scale-based accountability, not pouring practice. Here’s why: muscle memory is real, but bar noise, distraction, and custom (doing what the previous person did) will beat it every time. A barstaff member genuinely trying to pour 25ml will average 27–28ml. Acceptable. But ask them to do it 50 times in a 6-hour shift while taking orders and dealing with customers, and it drifts to 31ml. You lose money without anyone being dishonest.

The fix: weigh every open spirit bottle every week. A 70cl bottle of whisky that’s supposed to provide 28 measures (at 25ml each, plus 0.4ml for the glass) weighs roughly 770g full. If it weighs 650g, that’s 120g gone, which is roughly 15 measures. Check the till for how many whisky measures rang through. If it says 18, you have a 3-measure over-pour. That’s the number to address in training, not accusations of theft.

I use basic kitchen scales (£15 from Argos). Same scales, same spot on the bar, weigh every open spirit bottle on Tuesday morning before we open. One person does it. Two-minute job. The consistency is what matters — if you swap scales or personnel each week, the numbers become noise again.

Till reconciliation: the missing piece

Most cellar management training stops at the count. Pubcos will teach you how to dip a cask and rotate stock, and that’s genuinely useful. But it’s only half the picture. The other half is whether what you’ve counted actually matches what the till says you sold. That gap is where real money goes.

Here’s what I mean: your dip on a draught line says you’ve poured 40 pints. Your till says 38 pints rang through. Where are the two pints? Maybe staff gave them away (comps that didn’t ring). Maybe there’s a till error. Maybe the dip was wrong. Maybe two came out flat and went in the sink. You won’t know unless you reconcile the same day. Three days later, you’ve forgotten what happened, staff have moved on, and the discrepancy becomes noise.

Till reconciliation requires the same-day cross-reference of physical count against EPOS data, because memory loss and staff turnover make old discrepancies invisible. I pull yesterday’s Z-read every morning, check how many spirits, draught, cider and bottle lines rang through, then match that against what I counted last week and what I’m counting this week. If the variance is tight (within 2%), I move on. If it’s wide (over 3%), I drill into which line it is, and whether it’s a counting error, a till error, or a real loss.

The system I use logs both the physical count and the till data in one place, so I can spot patterns. But you can do this with a simple spreadsheet: physical count in column A, till data in column B, variance % in column C. Once you’ve done it three weeks in a row, the real patterns emerge. You’ll spot which staff member’s shifts have higher variance. Which line is drifting. Whether the problem is volume or price.

Building a training culture with your staff

Cellar management training is useless if your staff don’t understand why they’re doing it. If you hand a barstaff member scales and tell them to weigh bottles, they’ll do it. But if they don’t understand that over-pouring costs the pub money, and that the pub’s profit is what keeps their job secure, they won’t internalise the discipline. They’ll resent it.

This is where I differ from corporate training. You don’t need a classroom. You need a conversation.

• Show them the maths. “A bottle of whisky costs us £12 and sells for £42. That’s £30 margin. If we over-pour by 3ml per measure, that’s £1.50 per bottle lost. You pour 30 spirits a night, that’s £45 a night in lost margin. On a five-shift week, that’s £225 a week. Over a year, it’s over £11,000 from one person’s over-pouring.” They’ll get it.
• Let them help count. Bring a new barstaff member down to cellar on Tuesday morning. Let them see how much is actually in the kegs. Let them see what a properly cleaned line looks like versus a gunky one. Most staff have never been in a cellar. They think beer appears magically.
• Train on portion control first, then accountability. Don’t put scales on the bar and say “stop over-pouring.” That’s accusatory. Train on correct pouring technique first — show them how to use a jigger, how to pour level, why it matters. Then introduce scales as a consistency check, not a surveillance tool.
• Make it visible. I have a A4 laminated card on the bar that shows the margin on each spirit. Vodka is £30. Whisky is £30. Rum is £28. One visible card teaches staff faster than a hundred training sessions. They can see that every spirit has the same value, so there’s no game of “I’ll over-pour the cheap stuff.”

Staff training is training to your numbers, not training to corporate benchmarks. If your pub’s actual average pour is 27ml (because of glassware and fast service), then train to 27ml, not the nominal 25ml. The goal is consistency, not perfection.