Sound variability describes how much a space’s sound level rises and falls over time. It's often the true culprit behind employee complaints about noise, rather than overall sound level. Variability is the difference between a room that's steadily quiet or loud, where people can usually focus, and one where sudden spikes from calls and side conversations keep pulling their attention away.
Here’s what the research says about why high sound variability disrupts concentration, why the average level hides it, and how to measure and reduce it in your office.
Sound variability is the gap between the quiet and loud moments in a space, sometimes referred to as noise climate or sound level fluctuation. Acousticians describe it with two statistical levels: LA90, the level exceeded 90% of the time (the background floor), and LA10, the level exceeded 10% of the time (the intrusive peaks from speech and activity).
Sound variability is measured as the difference between the two, LA10 minus LA90. Whatever the right average level is for a given space, the goal for any office is to keep sound variability as low as possible, so a quiet baseline isn't repeatedly broken by louder events.
Most of the swings in an office come from familiar sources:
Of these, phone calls are the distraction workers name most often. In Gensler's 2025 Global Workplace Survey, 45% of workers in post-pandemic workplaces said a coworker's call regularly distracts them at their desk, ahead of nearby in-person conversations (41%), foot traffic (28%), printers and other equipment (24%), and outside noise (21%).
When sound constantly fluctuates, it forces your brain to reorient to every rise and fall, drawing attention away from the task in front of you even when you're trying to tune it out. Psychologists call the resulting disruption to short-term memory the irrelevant sound effect, and research links how strongly it occurs to how much a sound fluctuates:
Even in offices that don't seem loud overall, sound variability can significantly impede productivity and concentration, especially in open-plan offices where heads-down work happens a few feet away from calls and impromptu meetings.
The average sound level, LAeq, blends quiet and loud moments into one number, so two rooms with very different patterns can share the same average. A coffee shop with a constant hum and a quiet library where something clatters every 30 seconds could report similar LAeq values while feeling nothing alike to work in.
LAeq is also an energy average, which means loud moments count for more than quiet ones. Half an hour at 50 dBA and half an hour at 60 dBA average to about 57 dBA, not 55. That's why acoustic comfort depends on more than loudness: a room can meet an average-level target and still feel impossible to concentrate in.
Sound variability is measured by tracking statistical levels alongside the average over time, then comparing the peaks with the background.
|
Metric |
What it tells you |
|
LAeq |
Overall A-weighted sound level, the headline loudness figure |
|
LA90 |
Background floor, the level the room rarely drops below |
|
LA10 |
Intrusive peaks from speech and activity, the spike level capped in WELL v2 Feature S02 |
|
LAmax |
The single loudest moment in a period |
|
Sound variability (LA10 minus LA90) |
How far the room swings between its background and its peaks |
LA10 and LA90 each describe a share of the time you measured, so they're only as representative as the measurement period behind them. A short reading only reflects whatever happened in those few minutes, whether that was a quiet stretch or a burst of activity.
To capture these readings accurately, you have to measure them continuously, since a one-time reading will inevitably miss many of the moments that drive a room's spikes, like a late-morning run of calls or a meeting letting out beside a focus area. Data collected over days and weeks shows when and where the swings actually happen, which is the starting point for reducing them.
Reduce sound variability by first measuring where and when the swings happen, then narrowing the gap between the peaks and the background. You can't fix what you don't measure, and variability tends to concentrate in specific zones and hours, such as desks beside meeting rooms.
Once you know where and when the swings happen, the patterns in the data point to the likely cause and corresponding solution:
|
What the data shows |
Likely cause |
What to try |
|
A wide gap between peaks and background, driven by a low floor (LA90) |
Too little steady background sound to cover speech |
Raise the floor with sound masking. Research across 21 open-plan offices suggests keeping background sound in open work areas at about 40-45 dB LAeq. |
|
Peaks (LA10) concentrated at desks near meeting rooms, break areas, or printers |
Noisy activities sitting next to focused work |
Rezone so noise sources sit away from quiet work, place meeting-room entrances away from heads-down areas, and add enclosed rooms or walls where needed. Taller desk partitions add little shielding and can lead people to speak louder. |
|
Peaks clustered at certain hours, such as a late-morning run of calls |
Phone and video calls taken at open desks |
Move long calls to phone rooms or enclosed spaces, avoid speakerphones at open desks, and keep meeting-room doors closed. |
|
Peaks spread across a large open area with hard surfaces |
Sound carrying and lingering across the space |
Add absorption to ceilings, walls, and floors, and have a professional acoustician confirm reverberation with an acoustic survey. |
|
Frequent short, loud spikes (LAmax) with a moderate LA10 |
Footsteps, doors, and shared equipment |
Add carpet where footsteps are loud, choose sealed hinged doors over sliding or pivot doors, and move shared equipment away from desks. |
Many of these fixes come from the U.S. General Services Administration's Sound Matters guide, which recommends treating the space in order: absorb first, then block, then cover with masking.
Sound variability depends on each office's layout, activities, and schedule, so your own data makes a better benchmark than any single universal threshold. Compare sound variability before and after each change to confirm it worked.
Because sound variability changes from hour to hour, continuous data is the only reliable way to find where and when it happens and confirm that a fix actually narrowed the gap. Kaiterra's sound sensor tracks LAeq, LA10, LA90, and LAmax every minute, giving you that picture for every space you monitor. Reach out to our team to see how it could work in your office.