John Gravatt is chief operating officer of HVAC performance platform company ECM Technologies. Opinions are the author’s own.
For years, facility managers and building owners have been told to move away from reactive HVAC maintenance. Don't wait for equipment to fail. Inspect it regularly. That advice still holds; however, for many operations today, it is no longer enough.
A quarterly preventive maintenance visit may confirm that a unit is running. Still, it rarely shows whether a system is drifting out of performance, consuming more energy than it should or showing early signs of mechanical stress that will lead to failure.
Meanwhile, reactive maintenance is getting more expensive. Facility teams face tighter budgets, limited skilled labor, aging infrastructure and equipment expected to run longer and harder than intended. Tariffs are adding another layer of uncertainty to the cost of HVAC equipment and components.
When equipment fails, the cost is rarely just the repair; it expands into emergency labor, temporary cooling, expedited parts, tenant complaints and lost productivity. That environment calls for the next step: a shift from proactive maintenance to predictive, performance-based asset management.
Preventive maintenance has a blind spot
Preventive maintenance remains essential. Filters, coils, belts, bearings, motors, refrigerant circuits and controls still require skilled technicians.
The limitation is that preventive maintenance relies on schedules and checklists. Completing those tasks doesn't tell you how well all system components are performing; a system can run reliably every day and still cost more to operate than it should.
Over time, condenser and evaporator surfaces accumulate oil fouling that a visual inspection will not catch. Heat-transfer efficiency drops, the compressor works harder and longer to hold the same setpoint while mechanical stress builds and energy consumption per ton climbs. None of that trips an alarm or fails a checklist. The unit runs, and the space stays comfortable, while the asset quietly gets more expensive to own.
Predictive maintenance moves that conversation upstream. It means using actual condition and performance information: diagnostics, ultrasound and vibration analysis, heat-transfer performance, run-time and energy trends and service history to determine what needs attention, when, and how serious the risk is.
The important part is not collecting more data; it is turning that data into action.
Age is not a diagnosis
This matters most when deciding whether to repair or replace. Age matters, but age by itself is not a capital plan.
Two units installed the same week can be in different condition 10 years later. Operating environment, maintenance history, installation quality and run time all shape remaining useful life.
Replacing on age alone spends capital that may not need to be spent. Waiting for failure creates the opposite problem: a major capital decision forced at the price and lead time the market offers at that moment.
Predictive maintenance is a capital strategy
Consider six systems approaching the age at which replacement traditionally enters the capital plan. Performance data may tell a different story. One may be degrading materially and warrant replacement. Two may have correctable issues that can be addressed at a fraction of replacement cost. Three may be running efficiently with years ahead of them.
What follows is not a yes-or-no decision on replacement but a sequenced plan: replace one, correct two, monitor three and revisit in twelve months with fresh data. That defers significant capital without adding reliability risk, because the deferral decision rests on measured condition data.
Facility directors recognize a second major benefit: a condition-data-backed deferral is defensible. A 12-month trend showing stable approach temperatures, stable power draw and a normal-range bearing sound signature is evidence. Predictive data informs and documents the decision.
Tariffs raise the cost of getting it wrong
That matters more as replacement costs become less predictable. Recent tariffs have created the potential for higher HVAC costs. HVAC equipment is not directly targeted, but China remains a major source of compressors, motors and control boards. The ultimate impact will vary by equipment type, component sourcing, manufacturers and supply chains, but facilities leaders should prepare for continued pricing uncertainty.
Facilities leaders cannot control market forces. They can control how well they understand the equipment they own, so when a deferred asset reaches the end of its life, the replacement is planned and budgeted rather than treated as an emergency.
The hard part is not the data
Predictive programs rarely fail because the technology does not work. They fail on delivery. Data gets collected and never reviewed. Diagnostics get run by a technician trained on the instrument but not the interpretation. A finding lands in a service report and never reaches whoever controls the capital budget.
An effective program needs technicians with the right diagnostic tools, the training to interpret what those tools show and the ability to translate findings into practical options for owners. It also requires consistency: the same measurement, the same way, at the same interval and, for acoustic data, the same location. Replacing a one-time snapshot with consistent trending gives you equipment and system data that is both reliable and actionable.
Start where the risk is greatest
None of this requires a technology overhaul. Start with the equipment where failure would carry the greatest operational, financial, safety or tenant impact. Baseline it from current condition, recurring issues, service history and run-time patterns. Then add trending capable diagnostics that have the ability to influence asset repair/replacement decisions.
More data is not automatically better. Effective programs define what gets measured, who reviews it, which thresholds matter and how the assessment of trending data feeds asset decisions.
With equipment costs rising, the facilities that understand their assets best will face the fewest costly surprises.
Because in today’s capital environment, knowing what you don’t need to replace is just as valuable as knowing what you do.
