Anonymised practitioner case study. The organisation, precise locations and dates are withheld to protect the client and its staff. Internal metrics are not published, and none has been invented in their place.
The client
An organisation occupying a leased compound in a Horn of Africa city, used as office and guesthouse, with a contracted guard force.
The problem
Leadership expected a physical-security report recommending walls, cameras and hardware, a capital request. The assessment was asked to establish what the compound actually needed.
The method: layered assessment against realistic scenarios
- Defined the scenarios worth planning for from local incident history: forced entry, opportunistic theft, civil disturbance nearby, fire, and a blast event at a neighbouring premises.
- Assessed in layers, approach and neighbourhood, perimeter, access control, internal spaces, safe area, and evacuation, rather than as a hardware checklist.
- Observed the compound at three different hours, including a shift changeover and after dark.
- Interviewed guards about what they had actually been instructed to do, and compared it with the contract.
- Tested rather than reviewed: attempted arrival without notice; checked whether the visitor log was completed; timed how long it took to close the gate.
- Ranked every finding by scenario impact against cost to remedy.
What the assessment found
The perimeter was adequate. The weaknesses were procedural: the gate stood open during deliveries, guard orders did not match the contract, the safe area had no independent communications, no one held keys out of hours, and window film was absent on the elevation facing the street, the single cheapest measure with the greatest effect in a nearby-blast scenario.
What changed
Most recommendations were procedural and implemented within weeks. Capital spend was limited to window film, a communications set for the safe area and improved external lighting. Cameras were declined: nobody was going to monitor them, and an unmonitored camera is a record, not a control.
Transferable lessons
- Assess against scenarios, not against a hardware list.
- Visit at shift change and after dark. Daytime tells you very little.
- Read the guard contract, then ask the guards what they were told.
- Do not buy a control nobody will operate.