Notes from the field.
Retail AI, loss prevention, GDPR posture, the operational realities of running estate-scale security. Written by the engineers and customer-success people closest to the work.
RSS feedWhen a facial recognition match is wrong: what your staff should do
Every matching system is sometimes wrong. That is not a reason to avoid the technology, and it is not something to discover on the shop floor. This is the written procedure to agree before you switch anything on, and the record to keep when it happens.
Five myths about retail facial recognition
Say "facial recognition in a shop" and most people picture a system that scans every face, checks it against some giant database, and decides who to throw out. That version deserves the bad press. It is also not how responsible retail facial recognition works. Here are five myths worth clearing up.
Retail facial recognition: how a watchlist match actually works
Most of the fear around facial recognition comes from systems that act on their own. Retail facial recognition, done responsibly, is narrower: it compares a face to a list you built, flags a possible match, and asks a person to check. Here is exactly what happens, step by step, with a 47-second walk-through.
AI loss prevention needs an assistant, not just a detector
Almost every pound spent on AI in retail security goes into one place: the camera-side detector that watches. It mostly produces more alerts, not more time. The bigger opportunity is a second AI on the operator's side of the glass.
The self-checkout shrink problem: what AI on CCTV can and can't fix
Self-checkout is the loss-prevention story of 2026. The honest version: most of the leak is a lane-design and process problem first, with AI on CCTV as a second line, not a silver bullet.
Is the shop worker assault offence in force yet? (checked August 2026)
Most coverage says the Crime and Policing Act 2026 made assaulting a shop worker a separate offence. It will, but it has not started yet. Here is what has actually commenced, what has not, and what to do in the meantime.
UK stock shrinkage in 2026: what the BRC numbers don't tell you
Headline shrinkage is up. The category split underneath it is the more useful read. What UK loss-prevention leaders should actually plan for over the next 12 months.
AI loss prevention software: a buyer's guide for UK retailers
Twelve questions that separate AI loss prevention software that works in a UK store from software that demos well in a deck. ICO, ISO, human review, cross-store, edge vs cloud, the things to ask before signing.
GDPR-safe face recognition in retail: how we built it
Face recognition on retail CCTV can be GDPR-safe, but only if you make the architecture decisions early. A walk-through of how QuantumEye separates embeddings from video, audit-logs every step, and keeps a human in every consequential loop.
Why 'human-in-the-loop' is non-negotiable for retail AI
The pressure to automate every loss-prevention decision is enormous. Here's why operators who resist it end up with safer estates, lower legal exposure, and, counterintuitively, faster operational tempo.
UK retail crime statistics 2026: what the BRC report actually says
The BRC Crime Report 2026 in figures, quoted rather than paraphrased. Violence and abuse down by about a fifth, physical violence flat, and 5.5m shop theft incidents. What the numbers support, and what they do not.
Retail loss prevention KPIs: what to measure instead of CCTV review time
'How long does it take a manager to find a clip?' is the wrong question. The right question is how long until a forward-ready evidence pack is in the right inbox. Two different problems.
Edge AI vs cloud AI for retail security
Why we run inference on the camera, not in the cloud. Latency, bandwidth, GDPR, and the operational cost of getting it wrong.
Vectors and video: why they shouldn't live in the same place
The architectural decision that determines whether your face-recognition system passes or fails its first ICO review.
Forecourt night shift security: lone working, turnover and what helps
Why staff quit night shifts, and what changes when the system is the second pair of eyes. A look at the forecourt sector's hardest operating window.
Retail security and store analytics: one platform or two?
Most security tools and most analytics tools are sold separately. Here's why the same cameras can do both, and why splitting them is usually a commercial choice, not a technical one.
AI shoplifting detection false alarms: how we tuned our thresholds
The default thresholds out of the box are wrong for UK convenience. Here's what we changed, and the data we used to change them.
Why a cross-store watchlist changes loss prevention (and what to measure)
Per-store watchlists are how loss prevention was done in 2010. Estate-wide watchlists are how it gets done in 2026. Here's what changes, and how to measure the difference.
Get the monthly brief.
One email a month. The post-of-the-month, the retail-trends summary, and one customer-success snippet. No sales pitches, no event invites. Opt out in one click.