Category: Safety

Apartment building security tends to get attention twice. Once when it is set up, and once when something goes badly enough that residents start emailing the Committee. In between it is usually invisible, which is the point, but also part of the problem. When invisible security stops working, no one notices until the incidents pile up.

We have spent the last few weeks working through two very different security situations at two Auckland buildings. Both taught the same underlying lesson. The model matters less than whether the model is actually being delivered, checked, and adjusted when the building’s circumstances change.

Apartment building security: a slow escalation

The first situation started as a noise complaint between neighbours. Talking it through with the resident surfaced a bigger pattern. Groups of high school students were entering the building most weekday afternoons, riding the lifts for entertainment, gathering in stairwells and car parks, occasionally smelling of drugs. Residents were feeling intimidated. Some told us they had stopped speaking up in lifts when students got in with them.

The initial response covered the obvious moves. A notice to residents. Direct conversations with individuals we could identify. Contact with the schools. A report to Police, who attended once. Behaviour improved briefly, then returned and escalated. On one afternoon there were two separate groups using drugs in stairwells at the same time as students loitering in the car park, lift lobby, and lifts.

At that point we brought in a static guard, weekday afternoons, initially for two weeks. Daily activity reports showed students being encountered on almost every shift and leaving without argument when asked. Resident complaints dropped significantly within the first week. There was still the occasional incident, a lingering smell of drugs on one day, a suspicious item found and photographed on a stairwell on another, but the pattern of intimidation was broken.

The lessons from that one.

Residents will tolerate a surprising amount before they formally complain. By the time you hear about it, it has usually been going on for weeks. A single noise complaint is often the visible tip of something larger, so it is worth asking a few open questions rather than treating each report in isolation.

Police attendance is a one off tool, not a plan. In our experience, Police will show up when called and will disperse a group, but they do not have the resource to establish a presence at a private building. Anything requiring a repeat presence has to come from somewhere else.

A visible, uniformed adult in the right place at the right time is remarkably effective against opportunistic antisocial behaviour. The students were not there to cause serious harm. They were there because the building was easy and unsupervised. Removing the unsupervised part removed most of the reason to be there.

Two weeks is often enough to reset the pattern, but only if you review honestly at the mid point and are willing to extend if the outcome has not landed. We ran a mid point review with the approver and confirmed the guard was achieving what was intended before letting the second week run.

Pattern two: a model that stopped delivering

The second situation was a mixed use building that had moved earlier in the year from a static on site guard, running 86 hours a week, to a mobile patrol model. Four short patrols a day at randomised times within set windows. On paper the change was reasonable. Static guards are expensive, and mobile patrols cover more ground for less money.

What surfaced over a few weeks was that the delivery did not match the promise. A tenant reported an unauthorised person on site during a patrol window, but the person was still there well after the patrol was supposed to have happened. The activity report for the next day showed no logged checks for two of the four scheduled windows. A separate call to the security firm’s after hours line went unresolved and, oddly, ended with the operator asking for payment.

We escalated all three points to the provider, who ran an internal investigation and confirmed a mixed picture. One of the missed patrols was a logging error. The guard could not sign in to the site’s run sheet and recorded the visit against the wrong site. The earlier missed patrol, however, was genuinely missed. The call handling issue was escalated to their CEO.

A few days later another incident occurred. An individual on site through the middle of the day, moved on only after a tenant approached them directly. The tenant asked the Committee to consider switching providers and gave a five day deadline. The security firm’s own report of the same incident told a fuller story. Their guard had already approached the individual, who was cooperative and packing up to leave, when the tenant arrived and became verbally aggressive, which is what actually escalated things.

Both accounts were partly true. Neither on its own was the whole picture.

The lessons from that one.

Daily activity reports are worth reading, not just filing. Two consecutive days of gaps in the same patrol windows should trigger a question the same week, not the next month. Small anomalies compound.

When residents or tenants report an incident, get the security provider’s incident report of the same event before drawing conclusions. In this case the two accounts differed in ways that changed the recommended response entirely. From “the provider is not showing up” to “the provider is showing up, but the reporting infrastructure and the tenant provider relationship need work.”

Residents intervening directly is a signal that the reporting path is not clear enough. If a tenant felt the only option was to confront an individual themselves, then either they did not know who to call, or they had tried and it had not worked. Either way, that is fixable, with clearer signage, a tenant facing contact card, or a shorter escalation path, and it should be fixed before the next incident.

Cost is a legitimate reason to change a security model, but the model needs to survive contact with reality. Mobile patrols work well in some buildings and poorly in others. The variables that matter are things like how visible the building is from the street, how many entry points there are, whether there is a concierge or facilities presence during the day, and whether the tenant mix creates predictable friction points. When those variables change, the model should be reviewed.

What we take from both

A few things transfer between the two situations, and probably to most others.

Security is only as good as the feedback loop around it. If the Committee does not see incident data regularly, and the residents do not have an obvious path to report things, and the provider does not get told when something goes wrong, then whatever is happening on the ground drifts away from what was agreed at the AGM.

The right model depends on the building, not the industry. A retail adjacent mixed use tower has different needs from a residential only block near a school, and both change over time. It is normal to move between static, mobile, and hybrid arrangements as circumstances evolve. What matters is that the move is deliberate.

When behaviour escalates, act sooner than feels comfortable. Bringing in a short term static guard for two weeks is a small cost compared with the reputational and wellbeing cost of a building where residents feel unsafe.

Get the full picture before making the call. Every incident has at least two accounts, and often the useful decision emerges from comparing them rather than from either one on its own.

And finally, this is the least glamorous point but the most important one. The boring administrative work is what makes security actually work. Activity reports read, incidents logged, tenants briefed, mid point reviews held, providers held to their contracts. None of it is interesting. All of it is what stops the next incident from being worse than the last one.