Cellar management for pubs: the operator’s guide

Last updated: 26 June 2026

Most pub licensees discover their cellar management is broken only when the accountant shows them the numbers—or worse, when the brewery auditor finds the problem first. A 1% stock loss on wet sales quietly costs a typical pub £3,000–£5,000 a year, and it goes completely unnoticed because the loss is spread across dozens of lines, hidden in over-pouring, temperature waste, and the odd forgotten keg. You don’t need to be careless to leak money this way. You just need to treat cellar management like something that “sorts itself out.” It doesn’t.

If you’re running a pub on spreadsheets, guesswork, and the hope that your till data lines up with your physical stock, this guide is written for you. I built my own cellar management system from broken assumptions and lost thousands before it worked—and I’ve documented exactly what actually matters so you don’t have to.

Most pubs that move from a messy spreadsheet to a disciplined weekly count claw back 1–2 gross profit (GP) points within a couple of months. That’s not from finding theft. It’s from stopping the bleeding: tighter measure control, better cellar temperature, less waste from forgotten casks and over-poured spirits.

This article covers the systems, equipment, and weekly routine that will give you a stock variance number you can actually trust. No theory. Just what works.

Here’s why this matters: the number that actually matters is wet GP by line, not a single headline stock figure. Most operators focus on a monthly variance percentage and call it done. That tells you nothing. Spirits hide losses in over-pouring (a free-poured 25ml is often 32–35ml). Draught hides it in poor cellar temperature and bad line cleaning waste. And most stock ‘theft’ is actually measurement error and forgotten wastage. Once you know GP by line—cask by cask, spirit by spirit—you see where the real leaks are.

Why cellar management matters (and what most pubs get wrong)

When I took on my Marston’s pub, I ran stock on a tangle of spreadsheets and still lost track of partial kegs and spirit measures. I knew something was bleeding away, but I had no visibility into where. I’d get my brewery statement and spend an afternoon trying to reconcile figures that had already aged ten days. By then, the variance looked like a number I just had to live with.

The assumption most pub licensees make is that stock variance is either “part of running a pub” or the result of a dishonest member of staff. Neither is usually true. Spirits hide losses in over-pouring (a free-poured 25ml is often 32–35ml), draught hides it in poor cellar temperature and bad line cleaning waste, and most stock ‘theft’ is actually measurement error and forgotten wastage. A single spirit line over-poured by 5ml per serve, across 40 serves a week, is 200ml lost—about one bottle a month. Multiply that across your spirit range and you’re looking at real money.

The other blind spot: most pubs only check stock when the brewery auditor arrives or when they’re forced to do a stocktake for the accountant. That’s a quarterly (or worse, annual) snapshot. By then, the variance has compounded, and you have no idea where it came from. You’ve lost weeks of opportunity to tighten control.

The number that actually matters is wet GP by line, not a single headline stock figure. If your overall variance is 2%, but your cask ale variance is 0.5% and your spirits variance is 4%, you know exactly where to look. And once you know, you fix it.

The weekly cellar count routine

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. The routine takes about 90 minutes on a quiet morning, and it catches problems before they become losses.

Here’s the exact routine:

• Dip every cask and partial keg before service starts. Record the depth of liquid in each barrel. Use the brewery’s ullage table to convert depth to remaining volume. A dipstick costs £8–£15, and it’s the most important tool you own.
• Weigh every open spirit bottle (even nearly-full ones). Write the weight on the bottle in permanent marker at the start of each week. Weigh again the next morning before service. The difference is what was poured. A simple digital scale costs £20–£40 and eliminates 90% of spirit variance mystery.
• Count keg stock (empties and full kegs in the cold store). Compare to the previous week’s opening count, then subtract what you poured (from your dip sheet). This should match your till data for draught sales.
• Walk the spirit shelf and note any bottles that are missing, significantly lower than expected, or not accounted for in pouring records. Forgotten stock is a variance killer.
• Reconcile to till data the same day. Pull your EPOS draught and spirit sales from yesterday. Compare units sold (by till) to units poured (by weight and dip). If there’s a gap, find it now while you remember what happened.

At my pub, I do this every Monday morning. It takes 90 minutes. The first week, I found three partial kegs that had been sitting in the cold store for over a month—dead weight, assumed finished but never removed. A forgotten 12-bottle case of whiskey in a back cupboard. And my front-of-bar spirit measures were out of sync (one optic was stuck and overshooting by 3ml). Nothing criminal. Just invisible leaks.

StockTap pub stock app can automate the dip sheet and weight log (it stores your baseline weights, calculates the daily difference, and syncs with your till data), but the discipline of actually doing it—not just the tools—is what changes the numbers.

Equipment you actually need

You do not need an expensive system or specialist equipment to run tight cellar management. You need three things: a way to measure depth, a way to measure weight, and a way to record the data so you can reconcile it.

Minimum setup:

• Dipstick: £8–£15. Get one calibrated for your cask size (22.5 barrel for draught bitter, 9 gallon for keg lager). The brewery often provides these free if you ask.
• Digital scales: £20–£40 for a 5kg or 10kg capacity scale. Accuracy to 10g is fine. Keep it in the cellar and use it for every open spirit bottle.
• A notebook or spreadsheet: Write down the dip depth and spirit weights as you go. Don’t rely on memory. At the end of the week, convert dips to litres using the brewery’s ullage table and calculate spirit poured from weight difference.

Do not spend money on a “system” until you’ve proven that the discipline itself works. I’ve seen licensees buy £3,000 software only to abandon it because they were never going to do a weekly count anyway. The tool doesn’t matter if the habit doesn’t exist.

That said, if you’re already doing the weekly dip and weigh, moving that data into a simple app (rather than rewriting spreadsheets) saves time and reduces transcription error. SmartPubTools includes cellar management tracking built on the same logic I built by hand.

Temperature, line cleaning and draught waste

Cellar temperature is the silent cost killer in draught management. Most publicans assume that if the beer comes out cold, the cellar is fine. That’s wrong.

Beer should be stored at 12–13°C for real ale (cask). Lager and keg products typically sit at 4–5°C. If your cellar is running at 14–15°C, your real ale will go off faster, waste more product to sour casks, and pour with too much head (wasting volume). If your temperature swings—warm during the day, cold at night—you get condensation inside the lines, which promotes bacterial growth and forces you to run line cleaning more often, which means more disposal and more beer loss.

Get a simple max/min thermometer (£10–£15) and check it every morning. Write the reading down. If you’re regularly over 14°C, you have a refrigeration problem that’s costing you real money in waste and shelf-life loss.

Line cleaning is the other big variable. Most pubs should clean lines every 7–10 days. If you’re cleaning every 14 days to save caustic and time, you’re losing more beer to quality issues (cloudiness, off-taste, flat pints) than you’d spend on extra cleaning. Every time you clean a line, you lose about 2–3 pints of product down the drain. That’s factored into your variance calculation—it’s not stock loss, it’s operational waste, and it’s legitimate. But if you’re not recording it, your variance looks worse than it is.

Document when you clean every line and which caustic/cleaner you use. This gives you two things: (1) a record to justify “unexplained” draught variance to the brewery, and (2) evidence of whether you’re under-cleaning or over-cleaning based on the pattern of complaints and waste.

I run a log on a clipboard in the cellar. Date, time, line number, what I used, any notes (e.g., “tap was slow—maybe sediment”). Once I could see the pattern—I was cleaning one line every 6 days and it still tasted off, so I switched to a different cleaning regime—the problems cleared up and my draught GP stopped looking suspiciously high (which meant I was wasting less trying to compensate).

Spirits: measuring, over-pouring and variance

Spirit variance is where most pubs leak the most money, and it’s almost never theft. It’s measure control.

A free-poured 25ml is regularly 32–35ml. An optic measure that’s worn or stuck can drift by 3–5ml. A staff member who pours a “generous” measure because they’re generous (or because they’ve been told to) adds up to one extra bottle per line per week. None of these are dishonest—they’re just invisible.

The fix is brutal in its simplicity: weigh every open spirit bottle at the start of the week, write the weight on the bottle, weigh it again the next morning, and calculate what was poured. If you have a bottle at 850g on Monday morning and 620g on Tuesday morning, you poured 230g of spirit (roughly 11 25ml servings, or 10 if your pours are loose). Compare that to your till data: if the till says 15 spirits sold, you have a 5-serve gap. That’s your over-pouring, or a missed ring, or (rarely) a genuine loss.

Once staff know that every bottle is weighed and reconciled daily, over-pouring stops within a week. I’ve never caught a thief this way, but I’ve caught dozens of unintentional measure drifts.

Most spirit variance is not theft—it’s measurement error and over-pouring. Fix the measure control and your variance percentage drops by half.

The exceptions: bottles that go missing entirely (not a pouring variance, a genuine loss), spirits that don’t make it to the shelf (delivery variance), and occasional staff members who are stealing (rare, but real). The weigh-in system makes all three obvious: if a bottle is at 850g on Monday and you poured 230g, the bottle should be 620g on Tuesday. If it’s 300g, something’s wrong. That’s when you have a conversation.

Reconciling stock to till data on the same day

The final piece of the cellar management system is reconciliation. This is where the rubber meets the road.

Every morning, after you’ve dipped, weighed, and counted, pull your EPOS data from yesterday. For draught: compare the litres of beer you poured (calculated from dips) to the number of draught pints your till says you sold. Convert and reconcile. For spirits: compare the ml you poured (from weights) to the spirits your till says you sold. For bottled beer: count the empties and compare to till sales.

If they match (within a tight tolerance—5% for draught, 2% for spirits), your system is working and your variance is real and acceptable. If they don’t match, you find the discrepancy now while you remember the night before. Did you forget to ring a cask? Is the till price wrong? Did someone give away a drink and not void it? Find it immediately, not three weeks later in an audit.

This single step—reconciling the same day—is why my weekly variance went from a mystery to a number I could trust. I wasn’t doing anything differently. I was just catching problems before they aged and became impossible to trace.

At my pub, I do this on a simple spreadsheet (three columns: line name, poured yesterday, sold yesterday, variance %). It takes about 10 minutes. Most of the time, variance is 0–1%. When it’s not, I know immediately which line to investigate and which shift to ask about.