Here is the uncomfortable part. Theft from your shop is almost certainly happening right now, and if your till only adds up totals and opens the drawer, you have no way of knowing. You cannot manage what you cannot measure, and a basic till measures money taken, not stock lost. The gap between those two numbers is shrinkage, and for most independent shops it is completely invisible. This article is about making it visible.
Speak to us about retail software · +44 7494 618 651 · Mon to Fri, 9am to 6pm
Retail crime in the UK is not a rounding error. According to the ACS Crime Report 2024/25, independent convenience retailers lost £316m to retail crime in a single year. The same report records 6.2m theft incidents in a year, up from 5.6m the year before. Work that out and it is over 600 thefts every single hour, somewhere across the sector, while shops are open. The ACS describes the combined cost of crime and prevention as the equivalent of a 10p "crime tax" added to every single transaction a shop rings up.
Retailers are not standing still. The ACS Crime Report 2024/25 found that members spent more than £265m on crime prevention in the same year: cameras, shutters, alarms, fog cannons, the lot. That is a vast amount of money spent guarding against a problem, and yet the single most useful thing a shop can do, which is to measure exactly what is going missing and where, is the one thing most independents cannot do at all.
Why not? Because measurement needs data, and most independent shops do not collect it. The ACS Local Shop Report 2024 counts 50,387 convenience stores in the UK, with 71% run by independent retailers rather than the major multiples. The Retail Data Partnership estimates that around half of these stores use no EPOS at all: a basic till, a paper book or a spreadsheet. A till like that records that £4.20 went into the drawer. It does not record that the £4.20 was a meal deal, or that the energy drink in that meal deal was the fourth one to leave the shelf today when only two were scanned.
This is the heart of it. Shrinkage is the difference between the stock you should have and the stock you actually have. To see it, you need three things working together: a record of what you bought in, a record of what you sold, and a count of what is left. A simple cash till gives you none of these at item level. It gives you a grand total at the end of the day, which tells you nothing about which products vanished without being paid for.
Consider the everyday reality. A case of 24 lagers comes in. Over the week the drawer takes money for, say, 19 of them. Five are gone. Were five sold and miskeyed as something cheaper? Were five lifted off the shelf? Did two break in the stockroom and three walk out the door? On a basic till every one of those stories looks identical, because the only thing the till knows is the cash that arrived. The loss is real, it is recurring, and it is silent.
The moment every sale is recorded as a specific product, not just a price, the fog lifts. The shop's own demo system already does this as standard: scan or type the item, take card or cash, the receipt prints, and the stock count for that exact product decrements automatically. Multiply that across every transaction and you have something a basic till can never produce: a live, truthful picture of what left the building.
From that single foundation, loss becomes measurable in several distinct ways.
This is the big one. If the system knows you received 24 lagers and sold 19, it expects 5 on the shelf. When a stocktake finds 2, the system flags a discrepancy of 3 against that exact product line. Run that across the catalogue and the shop stops guessing. Instead of a vague sense that "we seem to lose money on the drinks fridge", the owner gets a ranked list of which specific lines are walking, by how much, and how often. That is the difference between suspicion and evidence.
When each member of staff logs in to the till, every sale, refund and void is attributed to a person. Nobody likes to think about it, but a meaningful share of retail loss happens behind the counter, not in front of it: false refunds, voided sales with the cash pocketed, "sweethearting" where a friend's basket never gets fully rung through. With per-staff tracking, a pattern of refunds clustered on one person's shifts is not an accusation, it is a number you can look at calmly and investigate. Without it, you will never even know to ask.
Not all shrinkage is theft. A great deal of it, especially in convenience, is perishable waste: bread, milk, fruit, chilled food past its date. The demo system tracks expiry on perishables and lets staff log waste and breakage explicitly, so spoilage is recorded as spoilage rather than disappearing into the same black hole as theft. That matters, because the fix for waste (order less, rotate better) is completely different from the fix for theft (reposition stock, change layout, supervise a till). If you cannot tell the two apart, you cannot fix either.
The same engine that exposes loss also exposes opportunity. The demo's reports show margin per line, top sellers, non-movers (dead stock that ties up cash on a shelf), busiest hours, waste and loss, and VAT broken out by rate. Loss prevention is not a separate product you bolt on. It is a natural by-product of finally knowing what you sell, item by item.
The pattern repeats across an independent shop: the costs you can see are rarely the ones that hurt. The till on the counter is visible. Shrinkage, dead stock and the slow drift of margin are not, which is exactly why they go unmanaged. A bespoke system that records every sale at item level turns those invisible costs into numbers on a screen, and that is the first step to doing anything about them.
Card-processing fees are another quiet, measurable leak. On the providers' published rates, Square and Zettle charge around 1.75% on every card sale while Revolut's published rate is 0.8% plus 2p, a gap that on an illustrative £15,000-a-month shop works out at over £1,500 a year. A system you own lets you choose your own card processing instead of paying whatever your till vendor bundles in. We break the full numbers down in the deep dive on shop card payment fees.
Margins are being squeezed from the other side too. Under the National Minimum Wage (Amendment) Regulations 2026 on legislation.gov.uk, the National Living Wage rose to £12.71 per hour on 1 April 2026, a 4.1% increase from £12.21. For context, the ACS records that the same rate was £11.44 back in April 2024. And the headline rate understates the real burden: research from the University of Stirling, working with the SGF in 2024, put the true, fully-loaded cost of employment at about £15.39 per hour once on-costs are included. That figure predates the 2026 rise, so the real number today is higher still.
When labour costs this much, the ACS found in 2024 that retailers absorbed the pressure however they could: 53% cut investment and 47% took lower profits. And the people running these shops are already stretched to the limit. The ACS, with the University of Stirling and SGF in 2024, found that 75% of convenience store owners and managers work 65 or more hours a week. In that context, money quietly bleeding out through unmeasured shrinkage is not an abstraction. It is the difference between a viable shop and an exhausting one.
Here is where the kind of system matters as much as the data it produces. Most of the ways a shop can get item-level data come with a catch. Epos Now advertises a system at £199, reduced from £849, but it requires a support and payments subscription from £54 a month plus card processing, on a 12 to 36 month contract. Symbol groups (and there are over 16,000 UK shops trading under a symbol fascia, according to The Grocer) often give EPOS free or subsidised, but it is tied to buying your stock from them. The data lives in someone else's platform, on someone else's terms.
A bespoke system from ESRE is built the other way round. You own it outright. There is no monthly software fee, no licence fee and no lock-in. If the relationship with us ended tomorrow, your system keeps running exactly as it did the day before. A core build starts around £2,000 to £3,000 as a one-off (larger scopes such as multi-site, a staff rota or an integrated website cost more, and those figures are illustrative). For an owner opening a second shop, a bespoke multi-site system at roughly £2,000 to £3,000 is materially cheaper than buying a second premium dealer till, which can run to £5,000 to £7,000.
One system replaces several disconnected tools: till, stock, ordering, reports and even your website can be parts of the same single source of truth. It is built in weeks, not months, it is noticeably fast, and it is built to last and grow as the shop grows. Because everything lives in one place, the loss reporting described above is not a premium add-on. It is just what happens when the till, the stock and the orders finally talk to each other.
Owning the system does not mean owning a toy. A bespoke shop build ships with a complete audit trail, automated backups, and data held on secure UK servers. A write-ahead log means the system can roll back to any second, so a fat-fingered stocktake or a bad bulk import is not a disaster. Full data ownership and export are standard, so the numbers are always yours to take. And the no-fault promise is simple: if something we built is faulty, we fix it at no cost, with no time limit.
The system is AI-ready by design. Since December 2025, AI tools have matured enough that your own team can help evolve the system in plain English, describing the change you want rather than writing code, inside failsafe environments with version control and automated backups catching mistakes before they reach the live shop. The catalogue is practical too: CSV bulk import, barcode lookup via Open Food Facts and phone-camera scanning mean nobody keys in 1,000 products by hand. Every smartphone becomes a scanner for deliveries, shelf work and stocktakes through a web app, with no app-store download, and the till is offline-capable so it keeps working if the internet drops and syncs later. Early clients also get free permanent hosting for a limited time.
For the full reasoning on why owning beats renting, see why bespoke, and for examples of what these systems look like in practice, see our projects.
Retail crime is a £316m problem and the ACS counts millions of incidents a year, hundreds every hour, plus a 10p crime tax on every sale. But the figure that should worry an independent owner most is the one nobody can quote, because it is the loss inside their own four walls that their till was never built to show. The shutters, cameras and alarms that the sector spends £265m a year on all guard the door. None of them count what walks out of it.
A bespoke system the shop owns closes that gap. Item-level sales, per-staff tracking, waste logging and stock-discrepancy reports turn shrinkage from a vague worry into a column of numbers you can act on. You cannot manage what you cannot measure. The whole point of owning the right system is that, finally, you can do both.
Speak to us about retail software · +44 7494 618 651 · Mon to Fri, 9am to 6pm