Twelve questions. Ask them before you hand over card details and you'll dodge the ways licensees get burned buying stock software: a counting method too crude to catch anything, maths that flatters you, or a contract built for a 40-site group when you run one pub. I've written out what a good answer sounds like for each, and the answers that should make you close the laptop.

I've sat through the demos. I've done the trials. I built StockTap because none of it worked the way my bar actually works, so yes, I have a horse in this race. The checklist below still applies to my product, and I'd rather you ran me through it too than bought anything blind. Every question here exists because I got stung by the thing it checks for.

The count: questions 1 to 4

1. How do I count a part-used bottle of spirits?

This is the fastest tell in the whole process. Most tools ask you to eyeball the bottle and call it "three tenths". Your eye is guessing, and the guess is always kind. A bottle you'd swear is a third full is often nearer a quarter, and over forty optics that drift is real money you can't see or prove.

The good answer is weighing. Bottle on a kitchen scale, the software knows the empty and full weights, and it tells you what's inside in millilitres. The walk-away answer is a tool that only does tenths and calls that precision.

2. What happens with a half-full keg and the head on a pint?

British Beer and Pub Association guidance says a pint served with a head only has to be 95% liquid. So up to 5% of every draught pint can be legal froth that never rings through the till as volume. A system that doesn't allow for that, or for line cleaning and pull-through, will report phantom losses every week and have you eyeing staff who've done nothing wrong.

Ask how it handles keg and cask calibration. If the person on the demo doesn't know what a dip reading is, they've never sold beer in Britain. My own take on getting this right is in the keg and cask calibration guide.

3. Does it work with no signal?

My cellar is a concrete box under the building. Yours probably is too. A cloud-only app that needs a live connection will die exactly where half your stock value lives. Offline counting with a sync when you're back upstairs is not a premium feature, it's the job.

4. Can I finish a full count on my phone, tonight, without training?

If the answer involves a laptop, an onboarding call and a PDF manual, you will do one count and never do another. The system you keep using beats the system with more features every time. A full count on a phone app should be doable in an evening; I've put real timings line by line in how long a pub stocktake takes.

The maths: questions 5 to 7

5. Are the margins calculated ex-VAT?

The single most common lie pub software tells is a GP% built on till prices with the VAT left in. It flatters your margin by around seven points, and you make ordering and pricing decisions off a number that was never real. Ask directly: is the sell price stripped to net before the GP calculation? If the demo can't answer, the answer is no. The full working is in how to calculate GP on drinks, and the same trap ruins stock valuations too, which I covered in ex-VAT versus inc-VAT stock value.

6. Is variance shown in pounds, per product?

A variance of "minus 1.2%" tells you something's wrong. A variance of "minus £212, and £160 of it is Grey Goose and Guinness" tells you what to do on Monday. You want the second. Percentages are for the report; pounds against named lines are for running the bar. What counts as a normal gap is its own question, and I've answered it honestly in what's an acceptable stocktake variance.

7. Does it still work without EPOS integration?

Integration demos brilliantly. But if the tool is useless until it's wired into your till, you're not buying a stock system, you're buying an integration project, and tied-house tills don't always play along. The good answer: you can upload or key in sales totals from any till and the variance sum still works, with integration as a bonus rather than a dependency.

The money: questions 8 to 10

8. Is the price published?

Check the website before the demo. Growyze publishes £79 to £139 a month per venue. Jelly publishes £129 flat. StockTap publishes its prices. Then there's the other lot: StockTake Online is entirely demo-gated, and the big procurement suites are all "book a call". A hidden price means the price depends on how keen you sound. I pulled every UK-relevant number I could find into what pub stocktaking software actually costs.

9. Is there a setup or onboarding fee?

WISK charges around $750 to get started on top of $249 to $399 a month. MarginEdge runs $350 a month per location. Those are US tools priced for US restaurant groups, and they count in ounces on top. For a single UK wet-led site, a setup fee in the hundreds buys you nothing a decent app shouldn't do free in an afternoon.

10. Is the price per site, per user, or per something else?

Per-site pricing is honest for a pub. Per-user pricing punishes you for letting your assistant manager count. And watch for tiering that holds basic reports hostage: if seeing your own variance history is a premium feature, the cheap tier is a brochure.

The exit: questions 11 and 12

11. What's the contract?

Monthly, cancel anytime, or it's a no from me. Enterprise suites still push multi-year terms, and 36 months is a long time to regret an afternoon's demo. Anyone confident in their product lets you leave. Simple as that.

12. Can I export my data as CSV whenever I want?

Your counts and your history are yours, not the vendor's. If you leave, that data trains your next system. If the answer to "can I export everything?" is woolly, you're not a customer, you're a hostage. Ask on the demo and make them show you the button.

The checklist on one screen

Screenshot this and take it into the demo. Pricing checked August 2026 from published vendor pages.
QuestionGood answerWalk away if
1. Counting part bottles?Weighing, to the millilitreTenths only
2. Draught and froth?Keg/cask calibration, line-clean allowance"What's a dip reading?"
3. No signal in the cellar?Offline count, syncs laterCloud-only
4. Count on a phone tonight?Yes, no training callLaptop plus onboarding project
5. Margins ex-VAT?Net prices throughoutGP% built on till prices
6. Variance in pounds?£ per product, worst firstOne percentage for the whole site
7. Works without EPOS link?Sales upload from any tillDead without integration
8. Price published?On the website"Book a demo" for a number
9. Setup fee?NoneHundreds up front
10. Priced per what?Per sitePer user, or reports paywalled
11. Contract?Monthly, cancel anytime12 to 36 month lock-in
12. CSV export?Everything, any timeWoolly answer, no button

What's deliberately not on the list

AI dashboards. Demand forecasting. Forty integrations. Supplier marketplaces. Some of that's useful eventually, and StockTap has some of it on the dearer tier, but none of it matters if the count itself is a fight. A pub that nails questions 1 to 6 with a £19 app is in better shape than a pub with a £350 suite nobody opens. If you want the head-to-head on named products, that's in the best pub stocktaking software in 2026, and whether any of it pays for itself is worked through in the payback maths.

The trial is the real demo

Whatever you shortlist, don't judge it on the sales call. Judge it on one real count. Load your top thirty lines by value, weigh the spirits, count the rest, and see whether you get a variance figure the same night without swearing at your phone. A trial that needs a card number before you can try is a red flag in itself; the free options and what they're actually worth are in what free stocktaking software really gets you.

Two counts a week apart tell you more than any demo ever will. One count is a photo. Two counts are a story.

Sources

  • British Beer and Pub Association — head-of-beer guidance, the 95% minimum liquid measure.
  • Published vendor pricing pages: Growyze, Jelly, WISK, MarginEdge, StockTake Online — checked August 2026.
  • Author’s own experience running a wet-led site in Washington, Tyne and Wear, and five independent liquor audits, December 2025 to July 2026.