It’s 02:47 on a Thursday in March. Your control room takes the call: intruder alarm active at a client’s site, a logistics warehouse on a small industrial estate just outside Reading. The closest mobile patrol is Dave, an SIA-licensed guard with twelve years on the job, currently halfway through a routine drive-through of a multi-storey car park about three miles away. Three minutes later he’s at the gates, beacons on. He gets out, walks the perimeter with his torch, checks the three main doors, tries the side fire exit, sweeps the rooflights, listens. Nothing. No forced entry, no broken glass, no signs of intrusion. Most likely a false trigger — they happen, particularly on a windy night like this one. Dave fills in the paper job sheet on his clipboard: on-site 02:50, off-site 03:18, result “no anomaly found”, signature. Back in the van, radios it through to the control room, carries on with the round.
The same morning at 09:15 your office phone rings. It’s the warehouse’s operations manager. Cold tone. “We had an alarm notification at 02:47 last night, but when our team arrived at seven this morning there was no sign that anyone from your patrol had been on site. No card left, no follow-up call, no email. Why are we paying for a keyholding and response service if your guard didn’t even bother turning up?” You know perfectly well Dave was there — you heard him yourself on the radio while you were in the control room — but on the client’s desk all there is, if anything, is a paper job sheet completed by hand, which the client has never seen, with a signature from your own employee. No photo of the perimeter at 02:55, no GPS timestamp on arrival, no incident report automatically delivered. Dave’s word against the client’s suspicion. And the client, in this game, is the one who decides whether to renew the £36,000 annual contract due in eight weeks.
This is the scene every UK security firm lives through at least once a month. It’s rarely outright dishonesty: sometimes it’s the night warehouse team not talking to day operations, sometimes it’s a facilities director who needs to justify the “manned guarding” line item to the CFO, sometimes it’s simply the fatigue of paying for a service that — when it works — is invisible. The dispute starts as a probe. If you don’t have objective, tracked evidence of the intervention, the client discovers it works. And six months later, at contract renewal, they ask for a fifteen per cent discount “because the service hasn’t always been up to standard”.
Why alarm response is the most fragile part of any guarding service
The structural problem with mobile patrol and keyholding is that the intervention almost always happens at night, on a deserted site, with no witnesses. The client is asleep. The night warehouse staff, if there are any, are in another part of the building. The guard arrives, walks the perimeter, and statistically eight times out of ten finds absolutely nothing. When the client turns up the following morning, they have no way of knowing whether the patrol actually arrived at 02:50 or at 04:15, whether the full perimeter was checked or just the main gate, whether the guard was on site for eight minutes or thirty. They see the same yard they left the night before. That’s it.
On that ground, raising a dispute is trivial. The client doesn’t have to prove the guard wasn’t there: they only have to suggest it, perhaps with a raised eyebrow at the contract review meeting. And you find yourself in the perverse position of having to prove a negative — that Dave really did arrive at 02:50, that the perimeter sweep really was carried out, that the twenty-eight minutes billed correspond to twenty-eight minutes on site. Without structured data, your only defence is a paper job sheet completed by hand, often without precise timings, with a signature from an employee the client doesn’t recognise — challengeable in five seconds: “How do I know that sheet was actually filled in at three in the morning, not in your office this morning?”.
You know the consequences. A single chargeback is small change: £320 on a routine call-out, £1,400 on a string of disputed responses through the month. But the real damage isn’t that single number. It’s the slow erosion of the annual contract. A mid-sized UK security firm with forty commercial and industrial clients loses somewhere between £30,000 and £90,000 a year to disputed invoices, forced discounts at renewal, contracts lost for lack of documentary proof, plus the time the control room manager and the operations director burn in defensive meetings justifying individual responses. Meanwhile the client’s insurers — the people who’d cover a theft if guarding had performed properly — are increasingly demanding objective digital evidence before settling a claim. Without that evidence they refuse the payout. The client loses the cover, and guess who they blame.
What it actually takes to certify a guarding intervention
To close the 09:15 call on the first attempt you need four elements, all in the same incident record, all automatic, none of them reliant on the guard’s willingness to fill in a form after two hours of walking around in the dark. The first is an explicit link between the alarm event received in the control room — with its unique reference, trigger time and the zone of the site that fired — and the response on the ground. Not a disconnected sheet, but a traceable chain: alarm 02:47 → patrol dispatched 02:48 → on site 02:50 → cleared 03:18 → back to base 03:25. A closed temporal thread, readable by anyone, exportable as PDF.
The second is GPS geolocation on arrival, captured at the moment the guard taps in on the company app, not self-reported by radio to the control room. If Dave arrived at the client’s address at 02:50, there has to be objective data showing coordinates, recognised address and timestamp, accurate to within a few metres. Same for departure: 03:18, same point, same GPS. The third is photos of the perimeter taken during the inspection, shot from inside the app, with timestamp and coordinates burned into the metadata and visible on the image itself: front gate secure, side fire exit intact, rooflights clean, loading bay empty. Three or four shots that unambiguously document what the guard saw on the ground. The fourth is the incident report completed on the guard’s tablet on site, with two distinct flows: “no anomaly” — standard checklist, outcome, optional notes — and “anomaly found” — description, additional photos, escalation to police via 999 where appropriate, alert to the client’s nominated contact.
