Trained pilots flying into cloud can be completely certain the aircraft is level while it is banking. The inner ear simply gets it wrong, and it gets it wrong with total confidence. That is why instruments exist, and why pilots are taught to believe the instrument over the feeling.
The lesson carries well beyond the cockpit: "this seems right to me" and "this has been measured" are two very different claims, even when the same person, in the same moment, is making both.
Now put that same gap in an airport, where the person making the call isn't judging a vintage. They're deciding whether an evacuation announcement can actually be understood.
The stakes, stated plainly
Hamad International Airport, Doha. North Node lounges. Public address across 30 zones, 205 speakers.
In most commercial spaces, a public address system exists to make announcements audible. Here, it exists to make evacuation instructions understandable in a hurry, under pressure, to people who may never have set foot in that terminal before and don't know where the exits are.
That's not a sound-quality requirement. It's a safety requirement. If people can't understand what they're being told, the system has failed, regardless of how good it sounds standing next to the speaker.
Why "it sounds fine" isn't enough
In an ordinary meeting room, "it sounds fine" is a perfectly acceptable standard. Nobody's safety depends on whether the projector audio is a little muddy, or a laptop mic clips on louder syllables.
Speech intelligibility is different. It's measurable. There are established ways to test whether an announcement can actually be understood in a given space, not just heard in it. And here's the uncomfortable part:
- A zone can sound perfectly good to the person standing in it while commissioning it.
- The same zone can still fail when it's actually measured.
- Large, hard-surfaced spaces — exactly the kind an airport is full of — are where this gap shows up most, because reverberation muddies speech in ways the ear doesn't always catch in the moment, especially a trained ear that already expects the system to work.
"It sounds fine" is an opinion. Intelligibility is a number. The two don't always agree, and in a life-safety system, only one of them is allowed to make the final decision.
What measuring 30 zones actually involves
This is the part nobody puts in the brochure, because it isn't glamorous. It's tedious, and it should be:
- Every one of the 30 zones gets verified individually. Not sampled. Not spot-checked. Each one.
- Zones get measured again as the space changes after ceilings go in, after furniture arrives, after the finishes that were still bare during the first fix are actually in place.
- Nothing gets signed off because someone stood in the room and thought it seemed alright.
That's supply-to-commissioning discipline, from first cable to final sign-off, applied to every one of 205 speakers. It takes longer. It costs more time on paper, and more patience in the field. It's also the only way anyone can honestly say the system works, rather than hope it does.
What actually gets caught
Not dramatic failures. Nobody finds a zone that's completely silent that would have been obvious from day one, to anyone. What careful measurement catches is smaller, and much easier to miss:
- A zone that reads perfectly well when empty, and noticeably worse once the space is furnished and occupied because soft furnishings and bodies change how sound behaves in a room that was tested when it was just concrete and glass.
- A gain structure that was set correctly at commissioning and has quietly drifted since, for reasons nobody can point to.
- A speaker that's technically working, drawing current, doing its job on a fault report but no longer covering the zone the way it was actually specified to.
None of these show up if the standard is "walk through and listen." All of them show up if the standard is "measure it, and prove it in writing."
Why this matters if you're the one signing off the project
If you're an owner or operator responsible for a space like this, here's the uncomfortable question worth asking before handover, not after an incident: who actually verified that this system works, and how do you know?
"The integrator said it was fine" is not an answer you want to be giving to a regulator, an insurer, or your own board, after the fact. A measured, documented sign-off is. That's the entire value of having someone independent doing the verification — not because the integrator is dishonest, but because nobody grades their own homework as rigorously as someone whose only job is the grading.
The transfer
Here's the part worth taking out of the airport and into an ordinary project.
Most rooms, boardrooms, meeting spaces, auditoriums are handed over on exactly the same basis that would be unacceptable in a life-safety system: someone stood in the room, it sounded alright to them, and that became the sign-off.
Nobody's safety depends on a boardroom microphone picking up the far end of the table clearly. But the discipline is the same discipline, and the value doesn't disappear just because the stakes are lower. A room that's been measured, not just listened to, is a room that works for the next person too, not just the person who happened to be standing in it on handover day.
Closing
In life-safety AV, "probably fine" is not a measurement. It's worth asking why we accept it anywhere else.
This project was delivered in a previous systems integration role, prior to AVYA's founding. It is shown as the delivery experience behind AVYA's independent consulting.
