Separating a Perception Complaint From a Real Fault
Why this matters
"It's just how they perceive it" is the most abused conclusion in the trade. It ends the call, it excuses the driveaway, and often enough to matter it is flat wrong: the fault was real, just intermittent or measured in the wrong place. Calling something perception is a diagnosis, and like any diagnosis it has to be earned. Earn it by ruling out a real fault first, or you will be back.
Perception is a diagnosis of exclusion
You do not arrive at "perception" by looking at the customer and deciding they are fussy. You arrive at it by eliminating the measurable causes one at a time until nothing physical is left. Treat it exactly like ruling out a diagnosis in any field: it is what remains after the real faults are gone, not a first guess you reach for to close the ticket.
The four screens before you call it perception
Run all four. Skipping one is how the real fault slips through.
- Reproduce it. Can you make the complaint happen on command? A fault you can trigger is a fault you can measure.
- Relocate the measurement. Move off the equipment to the exact spot the customer names. Source-perfect but point-of-use-weak is a real distribution fault, not perception.
- Watch it over time. In spec on a mild day can be out of spec under peak load or at a certain hour. Intermittent faults read clean between episodes.
- Check the sensory channel. The complaint may be a real side effect (noise, smell, flicker, vibration) that your primary gauge does not see. That is still a fault, just on a different axis.
The intermittent trap
The single most common way a real fault masquerades as perception: it only appears under a condition you were not present for. A weak branch under peak demand, a connection that fails when hot, a draw that outruns recovery. If the complaint has a pattern in time or load, believe the pattern over your one clean reading, and go capture it.
Tells that it really is perception
When these line up and the four screens come back clean, perception becomes the honest answer:
| Tell | Why it points to perception |
|---|---|
| It was always like this, they only noticed now | Nothing changed in the system |
| Only one occupant reports it, others in the same space do not | Individual sensitivity, not a shared measurable condition |
| It tracks their expectation or mood, not a load or a time | No physical correlate to chase |
| No reading, at any point or time, deviates | There is nothing to repair |
Never file a hazard under perception
Some complaints that sound subjective are early hazard signals: a smell near combustion equipment, a headache that clears when they leave, a tingle at a fixture, warmth at an outlet. These are not perception and not comfort. If a complaint could be a hazard, verify safety first (ventilate and shut down, or de-energize and verify dead), and only then continue the comfort diagnosis. A missed intermittent costs a callback; a missed hazard costs far more.
References
- Trade-standard practice on intermittent-fault capture and performance verification
- Manufacturer documentation on rated conditions and acceptable tolerances
- NFPA and OSHA safety basics for combustion and electrical hazard screening
- See related: The Comfort Complaint With No Measurable Fault; The Customer Feels a Problem Your Instruments Don't Show (decision tree)