The context
Surveillance investment across the Kingdom has been substantial, and the gap between what has been installed and what can be used in an investigation is wide. The reason is that surveillance is usually procured as a quantity of cameras rather than designed as an evidence chain, and the failure only becomes visible at the moment evidence is needed.
An industrial operator with a large perimeter, multiple gates and several hundred cameras had invested significantly and could not use the result. The cameras recorded. Retrieval took hours. Footage from the perimeter at night was, in practice, unusable — sufficient to establish that something had occurred, insufficient to establish what or who.
The operator discovered this the way most organisations do, during an incident investigation the system could not support.
The approach
ITBuilders assessed coverage, placement, lighting, retention and retrieval as a single chain, since a failure at any point makes the rest academic. Camera density was adequate. Placement was not. A significant proportion of the estate was aimed at areas that had been important when the site was smaller, and several pointed into direct sunlight for part of each day. Retention was configured uniformly at a period that satisfied nobody's actual requirement, having been set once at commissioning.
The redesign started from defined security objectives per zone. Perimeter detection, gate identification, internal area monitoring and safety observation are different problems requiring different camera types, placements and lighting. Treating them as one problem is why large estates underperform.
Identification points at gates were specified and positioned for facial and plate capture under the site's actual lighting conditions, including at night, which required lighting work as part of the security design rather than as a facilities afterthought. Perimeter coverage moved to a detection-first model with analytics tuned against the site's real conditions — wildlife, vegetation movement and weather all generate events, and analytics not tuned against them are disabled by the operations team within weeks.
Access control was integrated with surveillance so that a badge event and its corresponding footage are retrievable as a single record. This is the change that converts a surveillance system into an investigative capability.
Retention was set per zone against operational and regulatory requirement rather than uniformly, and storage sized accordingly. Network segmentation separated the security estate from corporate systems, with hardened device configuration and managed credentials, because a surveillance network is an attack surface and is frequently the least maintained one on site.
What changed
Incident retrieval now takes minutes and produces footage of evidentiary quality. The operator's security team spends its time on review rather than on search.
Continuity of operation
ITBuilders maintains the estate on an ongoing basis — camera health, analytics tuning as site conditions change seasonally, retention against current requirement, and the device hardening that keeps a surveillance network from becoming the softest target on the site. Cameras degrade, vegetation grows, and a system that is not maintained returns to unusable footage within a few years without anyone noticing until the next investigation.
