How Accurate Is a Keg Checker?

Last updated: 29 June 2026

Most pub licensees assume a keg checker gives them a precise reading of what’s left in the barrel — and then wonder why their stock variance still doesn’t add up. The truth is simpler and harder: a keg checker measures liquid depth, not loss. It tells you how much is in there right now, but it doesn’t tell you where the gaps went, whether you charged for every pint, or if your cellar temperature is wasting 3% on every cask to oxidation and line purge. If you’re relying on a keg checker alone to spot shrinkage, you’re already behind. This article covers what a keg checker actually does, what it misses, and why the pubs that claw back the most margin use it as one part of a system, not a substitute for one.

What Does a Keg Checker Actually Do?

A keg checker — whether an analogue dipstick, a dip tube with markings, or a digital probe — measures the depth of liquid in a barrel and translates that depth into an estimated volume. That’s it. It’s a measuring tool, not a loss-detection tool.

The keg checker works by inserting a marked stick or probe into the bunghole until it touches the liquid surface, then reading the volume directly from the markings or a digital display. On a standard UK keg (usually 36 or 40 litres), a decent dip will be accurate to within 1–2 litres under normal conditions — roughly 2.5–5% of the total volume. That sounds tight until you remember that a 2-litre margin on a 36-litre keg is £4–£8 in retail value, and that error compounds across eight, ten, or fifteen kegs.

Most pub operators assume a keg checker reading is final — they note the number, compare it to what they sold according till data, and call the difference shrinkage. But that’s where the real work actually starts, not ends.

The Real Accuracy Limits

Keg checkers have built-in accuracy limits that most licensees don’t account for:

• Temperature variance — A keg kept in a cold cellar at 8°C takes up less volume than one stored slightly warmer. Liquid contracts. A 1-degree variance can shift a reading by 0.5 litres. If your cellar drifts between 8–12°C across a week, your “accurate” readings aren’t comparing like with like.
• Keg wear and internal deposits — Kegs accumulate sediment and yeast. After six months of use, the internal surface of a keg is not smooth. A dip taken on Monday at the same liquid level will read differently than a dip on Friday if sediment has settled or shifted.
• Angle and technique — A dipstick held at a slight angle, or one that doesn’t reach the absolute bottom of the keg, will read high. Most pub staff dip kegs in a hurry. Inconsistent technique means inconsistent readings across staff, across weeks, and across kegs.
• Residual liquid after disconnection — A keg doesn’t empty completely. There is always head space and liquid trapped below the dip tube. A keg marked “2 litres” might have 2.5 or 1.5 litres usable. You can’t see the difference with a dipstick.

In practice, a good keg checker is accurate to within 5% under real pub conditions — not the 2% the manufacturer claims. And that 5% becomes invisible noise unless you’re reconciling readings against till data, staff count, and historical baseline numbers every single week.

What a Keg Checker Misses

This is where most pub operators get lost. A keg checker tells you the volume in a barrel. It does not tell you:

Over-pouring

A free-poured 25ml spirit measure is often 32–35ml. A keg checker cannot see a measure being poured. If your speed rack is pouring heavy and your tills are ringing up single 25ml measures, your stock will be short by 1–2 litres per week before any other loss is accounted for. This is the single biggest source of unreported loss in draught spirit and cocktail bars.

Cellar temperature and line waste

Draught lines need to be purged and cooled. Bad line cleaning wastes 2–4 pints per keg on average. Poor temperature control (warm cellars, blocked vents, unreliable coolers) causes oxidation and forces kegs to be declared waste before they’re technically empty. A keg checker reading doesn’t capture this. You have to measure it separately — temperature logs, line cleaning records, and waste declaration sheets.

Spillage and forgotten pours

A dropped pint, a drip tray that wasn’t totalled, a sample poured and not charged, a staff member’s shift drink — these are small amounts individually, but they compound. A keg checker won’t flag them because they’re already gone from the barrel. Your till won’t flag them because they were never charged. The gap appears as variance, and most operators blame theft or assume the keg checker is faulty.

Charging errors

You poured the pint. The till recorded it. But it recorded it at the wrong price, in the wrong category, or on the wrong till. Your stock numbers match your till numbers — but they’re both wrong. A keg checker won’t catch this because it’s not a stock problem; it’s a sales recording problem.

Why Weekly Counts Beat Single-Tool Reliance

I ran my first pub on a tangle of spreadsheets and an occasional dip of kegs when I remembered. I was losing track of partial kegs, spirit measures, and where the gaps actually came from. Stock was always short, variance was always a mystery, and I couldn’t make a decision about which lines to drop because I didn’t know which ones were actually losing money.

What changed was discipline, not a gadget. I built a simple weekly count routine: StockTap pub stock app later automated this, but the logic was manual first. Every Monday morning, I dipped every cask and partial keg, weighed every open spirit bottle, and reconciled the numbers against till data from the previous seven days — all on the same day, in the same spreadsheet. Within a fortnight, variance went from guesswork to a number I could trust and explain.

The number that actually matters is wet GP by line, not a single headline stock figure. Spirits hide losses in over-pouring. Draught hides it in cellar temperature and line cleaning waste. Most stock “theft” is actually measurement error and forgotten wastage. When you separate your count into line-by-line reconciliation — cask A is 2 litres short against its historical pull, cask B is perfect, cask C is showing 1 litre surplus because staff forgot to charge taster pours — you can spot patterns. You can see which kegs are consistently short. You can see which products bleed margin fastest. A keg checker reading tells you none of this because it’s a single-keg snapshot.

The pubs that move from a messy spreadsheet to a disciplined weekly count recover 1–2 GP points within a couple of months. Not because they stop loss overnight, but because they stop paying for the same loss twice — once as unknown shrinkage, once as wasted decision-making.

What Equipment Actually Works

You don’t need much, and you certainly don’t need expensive kit. Here’s what actually matters:

• A dipstick or dip tube — £5–£15. Make sure it’s marked in litres and fits standard UK kegs. Consistency matters more than cost. Use the same stick every week.
• A set of scales — £20–£40 for a decent digital set. Weigh every open spirit bottle weekly. The weight difference tells you exactly how much was poured, removed the guesswork from hand-measures.
• A temperature probe — £10–£25. Log your cellar temperature at the same time as your count every week. After four weeks, you’ll see whether your temperature is stable or drifting. Drifting temperature explains variance that a keg checker alone can’t flag.
• A paper count sheet or simple spreadsheet — Free. Record the date, the keg, the reading, the previous reading, the till pull for that product, the difference, and any notes (temperature, line cleaning, wastage declared). One row per keg.
• A till report — Free (it’s your own data). Pull a weekly sales report by product. Compare it to your stock count. The gap is your variance. If it’s tight and consistent, your system is working. If it’s wide and random, something is broken.

I’ve never recommended a smart keg checker or an automatic volume sensor to a licensee because they solve a problem that doesn’t exist — they make the reading slightly more convenient — while missing the actual problem, which is that no single reading means anything without context. SmartPubTools was built to handle that context. A keg checker measures. A system counts.

Building a Routine That Catches Real Loss

A keg checker is one input into a routine. The routine is what catches loss. Here’s what works:

Monday morning count (same day every week)

Dip every active cask and partial keg. Weigh every open spirit bottle. Log the cellar temperature. Record the date and time. This takes 20–30 minutes if you have six–eight active kegs and a reasonable spirit range. Do it before service if you can; the cellar is cooler and less disturbed.

Immediate till reconciliation

Pull your till report for the previous seven days. Compare sales by product against your stock reduction. If you sold 15 litres of bitter and your cask is only 14 litres short, where’s the other litre? (Line waste, spillage, taster pours, or till error.) If you sold 12 litres and the cask is 14 litres short, you’ve got 2 litres of unexplained loss — now you have something to investigate.

Line checking

Check your draught lines during the count. Look for leaks, blockages, or slow pours. Listen for pressure issues. A blocked line wastes beer; a leaking connection wastes money. Most pubs check lines twice a year. Check them weekly if your variance is wide.

Waste declaration

If you had to declare a keg waste, cut a line, or throw out spoiled product, record it. This is legitimate loss, not shrinkage. But it needs to be tracked separately so you can see whether waste is creeping up (sign of a cellar problem) or stable (sign of good practice).

Follow-up on gaps

If a line is consistently short by more than 3%, don’t wait. Check the line physically, audit till data for over-pours or ringing errors, review staff training on measures, and test the keg checker itself (dip the same keg twice to make sure technique isn’t the issue). Variance that you can’t explain is loss you’re still paying for.

This routine requires discipline but not much time. Thirty minutes a week compounds. Within a month, you’ll know which lines are truly profitable and which are bleeding you dry disguised as shrinkage. Most operators are shocked to discover they’ve been making decisions about which products to stock based on headline margin, ignoring the fact that a particular beer is losing 8% of its volume between keg and till.