Manned guarding or remote monitoring: how to protect a site, and at what cost

What each solution does — and does not do

Detecting is not acting

An alarm signals an event; it does not say what it is. Yet the vast majority of triggers are false positives: a cat, a door not properly closed, a tarpaulin flapping, a power cut. Without alarm verification, every alert wakes you up or needlessly calls out the police and gendarmerie, who in any case often require verification before they will travel out. That is where a human presence changes everything: a guard who checks, qualifies and reports the alert makes it possible to deal with the real one — and ignore the rest.

Conversely, a guard alone without remote monitoring cannot see what is happening at the other end of the site. Technology covers the ground, people provide the judgement. The two complement each other; they do not replace each other.

Measuring your real exposure

Before choosing, four questions determine the right set-up:

1. What needs protecting? High-value stock, plant and machinery, copper, fuel, data, or simply business continuity. A power-tool warehouse and a pallet store do not call for the same effort.
2. What is the layout of the site? Length of fencing, number of access points, blind spots, lighting, existing camera coverage. Four access points and no cameras call for more human presence than an enclosed, filmed site.
3. What are the vulnerable time slots? Nights, weekends, long weekends, the annual shutdown, periods of social tension, the aftermath of a break-in. Concentrating resources where the risk is real works better than spreading a permanent presence that is not much use.
4. What does the insurer require? Many policies make theft cover conditional on specific measures: a monitored alarm, fencing, lighting, sometimes manned guarding above a certain stock value. Reread the clause before deciding.

Three combinations that work

• A logistics warehouse in operation. Continuous remote monitoring and video; a fixed night post during loading periods, because the main risk is not outside intrusion but stock disappearing at the loading docks; patrols at weekends.
• An industrial site during its annual shutdown. Three weeks in summer: stepped-up presence, checks on any remaining deliveries, and above all routing technical alarms (refrigeration, fire, leaks) to a guard able to verify them and call the right technician.
• A vacant building or a site at a standstill. Temporary remote monitoring, two to four patrols a night at random times, checks on boarded-up openings, and a dog-handler team over the long weekends when theft and illegal occupation are most concentrated. See our guide protecting a construction site against theft.

CCTV: what the law requires

Cameras filming a place open to the public require prefecture authorisation (the video-protection regime). Cameras filming premises not open to the public fall under the General Data Protection Regulation and the Labour Code: staff and the works council must be informed, the purpose must be the safety of people and property, there must be no permanent surveillance of a single workstation, retention must be limited — generally one month at most — and a notice must be displayed for the people filmed. A security guard may be authorised to view the images for alarm verification; the recordings provide the evidence, the human presence provides the observation.

The classic mistakes

• Paying for a static post 24/7 when the risk is concentrated into forty hours a week.
• Stacking up cameras with no one watching them: video documents, it does not prevent.
• Patrols at fixed times: an intruder learns the pattern within two nights. Visits must be random and time-stamped.
• No alarm routing: an alarm that calls a number no one picks up at 3 am is useless.
• A guard with no instructions: what to do in the event of an intrusion, a leak, a fire starting, who to call, in what order. Everything must be written down before the first night.