Improve inventory accuracy at the bar

Last updated: 26 June 2026

Most pubs lose money every single week without knowing it. Not from theft — from measurement error. A 1% variance on wet sales quietly costs a typical pub £3,000–£5,000 a year, and the spreadsheet approach makes it invisible. The problem isn’t the spreadsheet itself; it’s that most licensees don’t have a repeatable, disciplined system to catch what’s actually happening in the cellar and behind the bar. You feel the margin squeeze but can’t see where it’s going. This article will show you exactly how to improve inventory accuracy at the bar, starting with a simple weekly count routine that catches stock loss within a fortnight. By the end, you’ll understand why the number that matters isn’t your total stock figure — it’s wet gross profit by line, and how to measure it in a way you can trust.

Why inventory accuracy really matters

The most effective way to improve inventory accuracy at the bar is to measure wet gross profit by line every single week, not a headline stock figure once a month. This is the insight that changes how licensees think about stocktaking. Your EPOS tells you what sold. Your stocktake tells you whether you made money on it. If they don’t match, something is wrong — and it costs you.

At my own pub, I was running stock on a tangle of spreadsheets and still losing track of partial kegs and spirit measures. The margin felt soft, but the numbers on the page said everything was fine. Then 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. That’s when I saw it: I was bleeding roughly 2% a month on spirits alone, almost all of it from free-pouring. A 25ml measure is supposed to be 25ml. Mine was coming out at 32–35ml. Once I had that data, I could fix it — and within eight weeks, I’d clawed back 1.5 GP points.

That’s not unusual. Most pubs that move from a messy spreadsheet to a disciplined weekly count recover 1–2 GP points within a couple of months. That’s thousands of pounds back in your pocket, without selling a single extra pint.

Why most pubs fail at stocktaking

Pub stocktaking fails for three reasons: no routine, no discipline, and no visibility of the right numbers.

First, most licensees stocktake once a month, which is too infrequent to catch trends. By the time you see a 5% variance, the damage is already done. A weekly line check takes 20–30 minutes and gives you instant visibility. If Tuesday’s count is 2% down from Monday, you know something happened between those two days. That’s actionable. A monthly stocktake is archaeological — you’re digging through four weeks of fog.

Second, the spreadsheet approach invites guesswork. You write down “three-quarter keg of Guinness” or “one bottle of Smirnoff.” You estimate the level. You forget to log the spillage from that dropped bottle or the staff drinks at the end of shift. By the time you reconcile against till data, the variance is so muddied that you can’t see where the loss actually happened.

Third, licensees focus on the wrong metric. They chase a headline stock figure — “I should have £8,000 in stock” — but they never ask: which line is actually bleeding? Spirits hide losses in over-pouring. Draught hides it in poor cellar temperature, bad line cleaning waste, and connection leaks. Most stock ‘theft’ is actually measurement error and forgotten wastage. Once you measure by line — spirits in bottles (weighed), draught in kegs and partials (dipped), soft drinks in crates (counted) — the pattern becomes clear.

The weekly line check: your foundation

A weekly line check is not a full stocktake. It’s a targeted count of your highest-value lines, reconciled against till data the same day. Here’s the structure:

Run it every Monday morning (or the first day after trading ends) before service. You’ll need 20–30 minutes, two people, and three pieces of information: opening stock (from last week’s count), what you sold (from the till), and closing stock (what you’re counting today).

The formula is simple:

Opening Stock + Purchases – Sales = Expected Stock. If Actual Stock ≠ Expected Stock, you have a variance. That variance is either measurement error (you miscounted or poured wrong), wastage (you logged it), or loss (you didn’t). The size and direction of the variance tells you where to look next.

Start with your top 10 lines by value: usually 5–6 spirits, 2–3 draught kegs, and soft drinks. Ignore the low-value stock for now; it muddies the numbers.

Spirits: weigh, don’t estimate

Open spirit bottles must be weighed on a set of digital scales. A 70cl bottle of Smirnoff at full strength weighs 875g. If it weighs 750g, you know you’ve poured approximately 125ml out of it — and you can tell the till how much sold. If the till says you sold 80ml, you have a 45ml variance. That’s real data. You can then ask: was that free-pouring, a promotional measure, or evaporation? Once you know, you can fix it.

Never estimate by sight. A three-quarter full bottle looks like 25ml is missing. It could be 50ml. Scales eliminate the guesswork.

Draught: dip every cask and partial

Draught beer hides more loss than anything else. You think the keg is half-full. It’s actually quarter-full because of a slow connection leak. You don’t see it; you think the customer just didn’t order Carlsberg that week. A dip stick costs £8 and gives you the exact level. Dip every cask and partial keg. Log the level. Do it again next week. If it dropped more than you sold, you have a leak or a temperature problem.

Cellar temperature matters here too. Ale at 55°F pours differently than ale at 62°F. If your cellar is drifting, your pint counts drift with it. A simple thermometer (£5) tells you if the environment is stable.

Soft drinks and ciders: count crates

Soft drinks variance is usually low if you count crates properly. Count full crates in, count empty crates out, weigh any open bottles. Most loss here is forgotten staff drinks or promotional pours — log them in your waste column.

Till reconciliation: same day

Run your till report the same day as the count. Don’t wait until Friday. Variance found on Monday is traceable; variance found on Friday is already forgotten. You need to know: sales by line (gross value), comps and staff drinks (by line), and net sales.

Then match it: if the till says you sold 45 pints of Guinness and your dip stick says the keg dropped 48 pints, you have a 3-pint variance (7% — acceptable for draught). If it says 45 sold and the keg dropped 60 pints, you have a 15-pint variance (33% — this needs investigation). That gap is real loss or measurement error. Find out which one.

Equipment you actually need

You don’t need much. You already have an EPOS and a till. What you need to add is measurement discipline:

• Digital scales (±1g accuracy): £15–25. Use for all open spirit bottles. Essential.
• Dip stick (metric markings): £8–12. Use for every cask and partial keg. Non-negotiable.
• Cellar thermometer: £5–10. Tells you if temperature is stable (should be 50–55°F for ale). Often overlooked, but it explains 20% of draught variance.
• A disciplined count sheet: Paper, spreadsheet, or — if you want to eliminate manual errors — a purpose-built app like the StockTap pub stock app.

That’s it. No expensive hardware. No overcomplicated system. The equipment is cheap. The discipline is what costs — but it’s the discipline that pays for itself in the first week.

Spirits, draught, and till reconciliation

Spirits hide losses in over-pouring because the variance is small and happens one pour at a time — a free-poured 25ml is often 32–35ml, which adds up to 40–50% loss over a month. Draught hides loss in connection leaks, temperature drift, and line cleaning waste. Soft drinks hide loss in forgotten staff drinks. Each requires a different counting method, but all require the same discipline: measure by line, reconcile against till data, and log variance the same day.

Here’s what a real weekly reconciliation looks like:

• Smirnoff 70cl: Opening weight 875g, closing weight 710g (165ml poured). Till shows 140ml sold at 25ml per measure. Variance: 25ml unaccounted (18% loss). Action: review pouring technique with bar staff.
• Guinness keg: Opening dip 85%, closing dip 45% (40 pints dropped). Till shows 38 pints sold. Variance: 2 pints (5% loss). Action: acceptable wastage from line cleaning and spillage.
• Carlsberg keg: Opening dip 70%, closing dip 20% (50 pints dropped). Till shows 35 pints sold. Variance: 15 pints unaccounted (30% loss). Action: investigate cellar temperature and connection integrity.

Notice the pattern: one line is fine, one is acceptable, one needs investigation. Without this breakdown, you’d see a headline “£200 variance” and have no idea which line to fix. With it, you’re precise.

The till reconciliation is crucial. If your EPOS isn’t feeding you line-level sales data, you can’t do this properly. Most modern systems do. If yours doesn’t, that’s the second problem to solve (your POS is the foundation of everything).

Common objections — and what actually works

Objection: “I don’t have time to stocktake every week.”

You’re right — you don’t have time for a full stocktake every week. A weekly line check takes 20–30 minutes with two people. That’s less time than you spend on admin in a single lunch shift. If you think you don’t have 30 minutes to save £3,000–£5,000 a year, you’re running the numbers wrong. That’s £60–100 per minute. Even accounting for setup time and learning, the payback is instant.

Objection: “My spreadsheet works fine.”

Spreadsheets work until they don’t. They’re vulnerable to formula errors, human typos, and the law of large numbers — small mistakes add up. More importantly, spreadsheets are slow. By the time you’ve entered data, reconciled against till, and chased variances, a week has passed. A structured count app eliminates manual entry error and gives you same-day visibility. If your spreadsheet is keeping you in the dark about loss, it’s not working — it’s just making you feel organised.

Objection: “Do I really need special equipment?”

You could estimate spirit levels by eye and guess keg depths by hand. You’d be wrong 20–30% of the time. For £40 total (scales, dip stick, thermometer), you’re right 95% of the time. The equipment isn’t a luxury; it’s the difference between guessing and knowing. Most pubs already have scales for weighing food. Use them for spirits too.

Objection: “Won’t the brewery stocktaker just do it?”

The brewery stocktaker comes once a month and counts your entire cellar to reconcile their deliveries against your usage. That’s a compliance check, not a management tool. It’s also not in your favour — if they find variance, it’s logged against you, and the conversation usually starts with “what happened?” You need your own weekly count so you already know what happened, and you can tell them why. That shifts the conversation from “you’re losing stock” to “here’s what I found and how I fixed it.” That’s the difference between being reactive and being in control.

Objection: “Is an app safer than a spreadsheet for my records?”

Yes. A spreadsheet on your laptop can be lost, corrupted, or accidentally deleted. An app stores data on a secure server. But more importantly, an app that’s built specifically for pub stock counting has validation rules built in — it flags impossibly high variances, reminds you of missing data, and keeps a history so you can see trends over weeks. A spreadsheet just lets you type whatever you want and won’t tell you if it’s wrong until you notice the margin has collapsed.

For more information, visit SmartPubTools.