The ticket comes in at 9:47am on a Wednesday. Third floor is too warm. The tenant has been dealing with it for a few days, they say, but finally decided to call. By 10:15am, a technician is walking the floor. By noon, there’s a diagnosis: the supply air temperature has been drifting for weeks. Nothing catastrophic. Nothing that tripped an alarm. Just a unit running slightly outside its normal range — long enough that the delta between what tenants expected and what they were getting finally crossed the threshold of tolerance.

The work order gets logged as a comfort complaint. It gets resolved in a day. And everyone moves on. What nobody records is the signal that existed six weeks earlier.

The data exists. In most operations, nobody is reading it — not because the team doesn’t care, but because the model requires someone to go physically check the equipment.

The Pattern Nobody Is Calling a Pattern

HVAC comfort complaints are the single largest category of reactive maintenance calls in commercial office buildings, according to BOMA research. And the majority of them — the temperature issues, the airflow problems, the humidity spikes — are not random failures. They are the visible endpoint of a condition that developed slowly, quietly, and measurably before anyone felt it.

A unit running 8% over its normal runtime to maintain setpoint is telling you something. A supply air temperature that has drifted 3°F from its baseline over four weeks is telling you something. A fan that is cycling more frequently than its historical pattern is telling you something.

And nobody goes to check equipment that hasn’t been reported as broken. That is the structural gap. Not a skills gap. Not a staffing gap. A visibility gap.

What IoT Condition Monitoring Changes — and What It Doesn’t

Runtime data from HVAC equipment is not new. BAS systems have been collecting it for decades in buildings that have them. What has changed is the cost and accessibility of deploying sensors in equipment that isn’t already connected — the RTU on the roof that’s been running since 2009, the AHU in the mechanical room that has no BAS integration, the split system serving a floor that was retrofitted after the original build.

IoT sensors deployed on those assets — monitoring runtime hours, supply air temperature differential, current draw, and cycling frequency — don’t make the equipment smarter. They make the anomaly visible before the tenant feels it.

ASHRAE Standard 180-2018 establishes minimum inspection and maintenance requirements for commercial HVAC systems, but inspection frequency alone doesn’t close the gap between scheduled visits. A quarterly inspection catches what is wrong at the moment of the inspection. Continuous monitoring catches what is developing between inspections.

The distinction matters because most comfort complaints don’t arrive at the quarterly inspection. They arrive on a Wednesday at 9:47am, three weeks after the quarterly PM was marked complete.

STAGE 1 Continuous Monitoring

IoT sensors track runtime hours, temperature differentials, and cycling frequency on legacy equipment continuously.

STAGE 2 Anomaly Detection

The system flags sustained deviations (e.g., a 12% increase in runtime) long before the tenant experiences discomfort.

STAGE 3 Proactive PM

A maintenance inspection is scheduled at a convenient time, preventing an emergency reactive complaint from ever occurring.

What Changes in the Operation When the Signal Exists

The facility team that has runtime anomaly visibility doesn’t wait for the tenant to call. When the unit on the third floor shows a sustained 12% increase in runtime over a 10-day window, the system creates a flag. The FM reviews it during the morning dashboard review — not in response to a complaint, but as part of normal operational visibility.

The work order that gets created is a PM inspection, not an emergency repair. The vendor visit is scheduled at a convenient time, not dispatched urgently on a Wednesday morning. The tenant never files a ticket because the temperature never drifted to the point of discomfort.

That outcome — a comfort complaint that never became a comfort complaint — doesn’t show up anywhere as a success. It doesn’t appear in any report. Nobody thanks the facility team for the problem that didn’t happen.

The True Value of Predictive Maintenance

According to McKinsey, organizations operating with predictive maintenance programs reduce maintenance costs 30-45% compared to reactive operations. That gap is not primarily about the repair itself. It’s about the labor, the urgency premium, the tenant relationship cost, and the deferred secondary damage that accumulates when a drifting condition isn’t caught early. Whether the operation has the infrastructure to hear the equipment is the only variable that mattered.

Sources: