Energy models are built to predict how a building should behave. Utility bills record how it actually behaves. When the two disagree, the difference is often called the performance gap, and in HVAC it can be significant.
Why models and reality diverge
A model is only as accurate as its inputs. It assumes design airflows, specified efficiencies, intended control sequences, a certain envelope quality and expected occupancy. Real buildings introduce variation at every one of these points.
Common causes of the HVAC performance gap
Duct and air leakage. If a meaningful share of fan airflow leaks out of ducts before reaching occupied spaces, the system must run longer or harder to meet the same load. The model rarely accounts for this unless leakage was specified and verified.
Controls that do not match the design intent. Schedules left on 24/7, setpoints overridden, economiser logic disabled, sensors out of calibration, and valves or dampers that no longer respond correctly can all raise consumption without anyone noticing.
Incomplete commissioning. When commissioning is compressed by programme pressure, issues that should have been caught at handover remain in the building: unbalanced airflow, incorrect fan speeds, simultaneous heating and cooling, and untested sequences.
Envelope performance. Infiltration through the building envelope changes the actual cooling load, particularly in hot climates where uncontrolled outside air adds both sensible and latent load. An envelope that is leakier than modelled shifts the burden onto the HVAC plant.
System degradation. Filters load, coils foul, belts slip, seals age and ductwork is disturbed during maintenance and fit-out. Performance drifts gradually, and without periodic measurement the drift is invisible.
Operational differences. Occupancy, equipment loads and operating hours often differ from those assumed in the model. This is a legitimate source of variation, but it should be separated from avoidable physical and control problems.
Investigating the gap
A useful investigation separates what is explained by operation from what is explained by physical condition:
• Compare metered consumption with the model, adjusted for actual weather and occupancy.
• Trend control data to check schedules, setpoints and sequences.
• Measure airflow at the fan and terminals against design.
• Test duct leakage to quantify delivered-versus-supplied air.
• Inspect the envelope and plant for obvious infiltration paths and degradation.
The aim is to rank causes by their likely impact, so that effort goes to the problems that will actually reduce consumption.
Why leakage matters more in hot climates
In a hot climate, supply air is expensive. It has been cooled and often dehumidified at significant energy cost. Air that leaks into a hot ceiling void or plant room is not only wasted; it is conditioned air handed to a space that is already hotter than the occupied zone. Return leakage is equally costly: it can draw hot, dusty air into the system, raising the coil load and the burden on filters. For buildings in regions with long cooling seasons, even moderate leakage can translate into a sustained energy penalty throughout the year.
The compounding effect of small deviations
The performance gap rarely has one cause. A system may have slightly leaky ducts, a control schedule that runs longer than intended, a fan running faster than necessary to compensate and an envelope that admits more outside air than modelled. Each deviation adds only a few percent, but together they can move consumption well away from the prediction. This is why chasing a single “big” cause often disappoints: the gap is the sum of many small ones.
A practical six-step approach for building owners
- Establish the baseline. Gather metered energy data, operating hours and, where possible, sub-metering for HVAC plant.
- Compare with the model or benchmark, adjusting for weather and occupancy so that you compare like with like.
- Review operation. Check schedules, setpoints, sequences and any manual overrides.
- Measure physical condition. Test airflow, duct leakage and key pressures, and inspect filters, coils and ductwork.
- Rank the findings by expected energy and comfort impact, not simply by how easy they are to fix.
- Correct and verify. Carry out the highest-impact corrections and repeat the measurements to confirm the improvement.
Data worth collecting
Good investigations are built on good data. Useful items include design airflow and pressure schedules, TAB and commissioning reports, control sequence descriptions, trend logs, utility bills over several years, and records of maintenance and modifications. Where these records are missing, a measured baseline becomes even more valuable, because it creates the reference that later comparisons depend on.
What not to do
Avoid assuming that new equipment will close the gap. If leakage, controls or envelope problems are driving consumption, a more efficient chiller or air handler will still be operating in a system that wastes part of its output. Treat equipment replacement as a conclusion drawn from evidence, not as the first response to a high energy bill.
Where the biggest savings usually sit
In many buildings, the largest and most persistent savings come from correcting issues that quietly waste energy every hour of operation: leaking ductwork, controls running outside their intended schedule, and airflow imbalances that force fans and plant to work harder than necessary. These issues rarely appear in a design review, but they are measurable. By quantifying them first, owners can prioritise corrective work where it will produce the most reliable improvement, and can track whether the savings actually materialise after the work is done.
Conclusion
The performance gap is not a mystery. It is the sum of many small, measurable deviations between design intent and real condition. Buildings that are measured, diagnosed, corrected and verified close that gap. Buildings that are only modelled keep paying for it.
