Marston’s Cellar Management: The Numbers That Actually Matter

Last updated: 26 June 2026

Most Marston’s tenants lose £3,000–£5,000 a year to cellar stock waste and don’t even know it’s happening. That’s a 1% loss on wet sales—which sounds small until you realise it’s sitting in your profit margin, bleeding away week after week. The problem isn’t that you’re a bad operator. It’s that you’re measuring the wrong things. You’re looking at a headline stock figure when what actually matters is wet GP by line, temperature variance, and whether your till data matches your physical count the same day. I spent years running stock on a tangle of spreadsheets and partial keg notes until I built a disciplined counting routine around a dipstick, a set of scales, and a single number I could trust. Within a fortnight, the guesswork stopped. Here’s exactly how to do cellar management in a Marston’s pub—and why most operators get it wrong.

Why Marston’s Tenants Lose Money on Stock

If you’re running a Marston’s pub, you’ve inherited a tied estate model where your margins are already tight. The brewery sets the buy-in price. You can’t renegotiate the rent. The only margin you control is operational efficiency—and cellar stock is where most of that margin disappears.

The most effective way to identify hidden stock loss is to measure wet GP by line rather than watching a single headline stock figure, because losses hide in over-pouring, temperature variance, and forgotten wastage rather than a dramatic shortage. This is the insight that changes how you count. A spirit bottle that looks full at the till might actually be worth 15% less if your free-pour is running 32ml instead of 25ml. A draught line that runs 2°C too warm is generating invisible waste through poor head control and line cleaning loss. A keg that sat for three days after a partial pour and wasn’t recorded is gone from your count, but not from your costs.

Most Marston’s operators count stock monthly or quarterly, reconcile the number against a spreadsheet, and assume the difference is either theft or acceptable shrinkage. Wrong. By the time you see a monthly variance, you’ve already lost 4 weeks of margin. The losses that matter are the ones you catch the same week they happen.

The Real Cost of Poor Cellar Counting

Let’s put a number on this. A typical Marston’s pub runs £15,000–£20,000 of wet stock. At a 40% GP target, you’re aiming for £6,000–£8,000 of profit on that stock over a year. A 1% loss—which is invisible to most operators—costs you £3,000–£5,000 of that. That’s not a rounding error. That’s your Christmas bonus, your emergency fund, or the reason your business barely broke even when the numbers said it should have made money.

Where does that 1% go? Spirits hide it in over-pouring (a free-poured 25ml is often 32–35ml when measured). Draught hides it in poor cellar temperature—run a keg 3°C too warm and you’re venting CO₂ and generating foam waste that never gets sold. Cider and wine hide it in partial bottles and forgotten opened stock. Most stock “theft” isn’t an employee stealing a bottle. It’s measurement error, forgotten wastage, and counting the same partial keg twice because you didn’t write down when you opened it.

The cost compounds because you don’t see it. Your till says you sold £500 of spirits yesterday. Your stock sheet says you should have 45 bottles of Absolut. You count 44 bottles. You assume one went missing. You write it off. You do this four times a month. By year-end, you’ve written off £3,000 of variance and called it theft. Most of it was over-pouring that you never measured.

Reconciling physical stock against till data the same day is the only way to catch which line is losing margin, because a one-week delay between counting and investigation makes it impossible to link a variance to a specific cause. This is the discipline that separates operators who know their numbers from operators who run on hope.

Weekly Line Checks: The Foundation of Stock Control

Stop counting stock monthly. Start checking every line every week. This takes 90 minutes once you have a routine, and it catches 80% of variance before it becomes permanent loss.

Here’s the actual process I use:

• Spirits: Weigh every open bottle on a digital scale. Record the weight. Compare it to last week. A 25ml pour should move the needle by about 25g. If a bottle dropped 200g and your till shows 6 sells, you have a measurement problem (over-pouring) or a till entry problem.
• Draught: Dip every cask and partial keg. Write down the reading. Measure cellar temperature (should be 12–14°C for ale, 8–10°C for lager). Check the line connectors for leaks. A leaking union will waste 2–3 pints a day without you noticing.
• Cider and wine: Count bottles, note the level in any opened bottles, and reconcile against till data for that brand.
• Reconcile same-day: Pull your till data for the same week and compare it to your physical count changes. If your till says 40 pints of Guinness sold but your keg dropped 45 pints, you have 5 pints of waste to explain (foam, line cleaning, overpour into a glass that wasn’t charged). If that happens every week, your line pressure is wrong.

The equipment you need is minimal: a StockTap pub stock app or a notebook, a dipstick (£15 from a brewery), a digital scale (£20 from Amazon), and a thermometer (£8). Total investment: less than £50. Time: 90 minutes per week. Return: you’ll claw back 1–2 GP points within two months once you identify which line is leaking margin.

I ran this routine at my own pub and the difference was stark. Week one, my variance was all over the place—I didn’t know what I was measuring. By week four, my weekly variance was within 2–3%, and I could tell you exactly which spirit was being over-poured and which draught line had a temperature problem. By week eight, my GP improved by 1.2 points. That’s £1,500–£2,000 a year at typical volumes.

Measuring Wet GP by Line, Not a Headline Figure

This is where most operators fail. They look at a single “stock variance” number and treat it like the health indicator for their cellar. It’s not. That number hides where the actual problem is.

What matters is wet GP by line. You need to know:

• Which spirit is generating the lowest margin (usually vodka and gin because they move volume and it’s easy to over-pour).
• Which draught line is losing the most pints to waste (usually a keg that’s run warm or a line that needs descaling).
• Whether cider and wine are selling at expected margins (many operators underestimate opening loss and forget to charge partial bottles correctly).

To measure this, you need three pieces of data: what your till says you sold, what your physical count says left the cellar, and what your cellar conditions tell you about waste (temperature, line condition, age of keg). If your till says 50 pints of Stella sold but your keg dropped 55 pints, and your cellar was 15°C (warm for lager), you have a 5-pint waste problem linked to temperature. Fix the temperature and you fix the waste. If you don’t measure it by line, you never find the problem.

Spirits hide losses in over-pouring, draught hides losses in temperature and cleaning waste, and cider hides losses in forgotten opened stock—which means you need a different measurement approach for each line. A single stock-take number won’t tell you any of that.

The discipline here is ruthless: measure by line, reconcile by line, and fix by line. Don’t average your variance across all spirit bottles. Don’t assume all kegs waste the same amount. Measure what’s actually happening, and you’ll find the 1% loss faster than any other operator in your pubco.

Common Cellar Mistakes (and How to Avoid Them)

Counting without recording the conditions. You count your stock on Monday morning and write down “Absolut: 48 bottles.” Useless. Write down the count, the date, the time, the bottle weight, and the till data for the same period. Otherwise you have no way to know whether the variance came from over-pouring, a till error, or an actual shortage.

Ignoring cellar temperature. This is a hidden cost killer. Run your cellar 2°C too warm and you’re venting CO₂, generating foam, and wasting 10–15% of your draught margin. Most Marston’s pubs have a cellar thermometer that nobody reads. Check it every week. If it’s wrong, you have a justification for variance that’s not your fault—and you have a clear fix.

Not reconciling the same week. If you count on Monday and don’t reconcile against till data until Friday, you’ve had four days of trading between count and investigation. You can’t link a variance to a cause anymore. You become a guessing game. Reconcile the same day. Yes, this takes discipline. Yes, it’s worth it.

Treating the brewery stocktaker as your quality control. The brewery stocktaker is there to verify your stock and raise a discrepancy if there’s a big problem. They are not there to help you manage your cellar. If you wait for the brewery count to find out you have a problem, you’ve already lost a month of margin. Do your own count every week. The brewery count is a sanity check, not a discovery tool.

Running an old spreadsheet instead of a proper system. Most Marston’s operators run stock on Excel or a notebook. This works if you’re disciplined. But most operators aren’t. They forget to enter a count, double-enter a partial keg, or lose the notebook. A simple tracking system—even a free one—removes the human error. SmartPubTools has built-in cellar tracking that links your count to your till data automatically, but even a basic Google Sheet where you record the same fields every week will beat a paper notebook.

Building a Routine You’ll Actually Stick To

The best cellar management system is the one you’ll actually use. Here’s how to build a routine that sticks:

Pick one day a week and do it at the same time. I do mine on Tuesday morning before the lunch shift. It takes 90 minutes, and I know it’s coming. I’ve carved out the time. I’m not squeezing it in between a delivery and a staff problem. If you try to stocktake whenever there’s a gap, you’ll miss half the weeks and your data will be useless.

Use the same equipment every time. The same scale, the same dipstick, the same thermometer. Different scales will give you slightly different readings (especially for spirits). If you use a different one each week, you’ll chase phantom variance.

Record the same fields every week. Brand, count, weight or dip level, date, time, till sales for that brand (or line), cellar temperature. This consistency is what turns data into insight. A casual count is just a number. A weekly record of the same fields shows you trends—and trends show you problems.

Share the routine with your team. If only you know how to count stock, it stops the day you’re ill or on holiday. Train your manager or a senior member of staff to do the count the same way you do. Give them the same equipment. Let them own the number. You’ll get consistent data and you’ll free up your own time.

I objected to weekly counting for years because I told myself I didn’t have time. I was wrong. I was spending two days a month on a crisis stock variance, chasing ghosts. A 90-minute weekly count costs me 6 hours a month and saves me the 16 hours I used to spend on variance investigation. The math is obvious once you see it.