Noise monitoring in open-plan offices — comfort without recording conversations
Open-plan noise monitoring without recording a word: the sensor measures decibels only, while thresholds and alerts guard quiet zones before complaints pile up.
Zespół Nextriv4 min read

Noise monitoring in an open-plan office carries a paradox: noise is the parameter office workers complain about most often, and at the same time the one almost nobody measures — because a "sound sensor" sounds like a wiretap. Yet measuring noise has nothing to do with recording. Only the sound level in decibels is registered: no content, no voices, no possibility of playing anything back. The result is that instead of anonymous complaints that "you can't work here", the facility manager gets numbers — and numbers, unlike opinions, can be compared, defended and improved.
Noise in the open plan — the problem no chart shows
Most office comfort parameters are already being measured by someone: temperature is watched by the building automation, air quality increasingly by CO₂ sensors. Noise remains a blank spot, even though in an open-plan space it is most often what decides whether focused work is possible at all.
The trouble with noise is that it is extremely subjective. For orientation: a whisper is about 30 dB, a calm conversation about 60 dB, and an open-plan office at peak hours can sustain a background for long stretches at which deep work becomes a struggle. Worse still, the decibel scale is logarithmic — a 10 dB increase is perceived roughly as a doubling of loudness. That is why the dispute of "it's loud here" versus "you're exaggerating" will never settle itself: one person sits next to the meeting zone, the other in a quiet corner, and both are telling the truth.
Without measurement, managing acoustics looks the way it looks: desks shuffled blindly, quiet-zone rules nobody enforces, and phone booths bought where there was space rather than where they were needed.
Decibels instead of recordings — how open-plan noise monitoring works
A measurement-grade noise sensor works differently from a dictaphone: the microphone serves solely to determine the sound pressure level, and only numbers reach the platform. There is no audio recording, no speech recognition, no content — not a single word can be reconstructed from the data. This is privacy by design: the difference between measuring and recording is technical and irreversible, so it can be communicated to the team outright, written into the office policy, and the wiretap topic closed before anyone opens it.
In measurement practice, three metrics matter and are worth understanding:
- SPL — the instantaneous sound level. It shows what is happening right now; useful for local signalling.
- Leq — the equivalent level, i.e. the averaged "dose" of noise over time. It is the best indicator of a zone's acoustic background and the basis for comparisons between places and weeks.
- Lmax — the maximum level. It catches the peaks: a slammed door, a burst of laughter, chairs being dragged. A zone with a good Leq but frequent Lmax spikes is a different problem from a zone that is constantly loud.

Nextriv Sense Noise is designed for exactly this kind of measurement: it reports SPL, Leq and Lmax simultaneously, measures across 30–120 dB with 0.1 dB resolution, and its fast 125 ms time constant captures impulsive noise. You can choose the A-weighting curve — close to the perception of the human ear — or C for loud environments. The flat, 20.5 mm enclosure blends into the wall instead of looking like lab apparatus, and the built-in LED can act as a local "light reminder" when a threshold is crossed — often its mere presence is enough for conversation levels to drop without any intervention.

Where to hang the sensors and which thresholds to set
One sensor for a whole floor measures an average nobody cares about. What makes sense is a map: one sensor each in the quiet zone, in the main part of the open plan, by the meeting zone and near the kitchen or the coffee machine — that is, in places with different expected acoustic profiles. The sensor goes up roughly at the head height of a seated person, away from sources that would skew the measurement towards a single desk. Installation is double-sided tape or screws and configuration is a tap of the phone (NFC) — no cables and no electrician, and the replaceable batteries last, depending on the reporting mode, even years. How to manage that resource deliberately is covered in our piece on sensor battery life.
Thresholds are best built on Leq, not on instantaneous SPL — a single laugh should not raise an alarm. In the Nextriv platform every metric gets up to four thresholds: two warning and two critical. A sensible starting point for the quiet zone is a warning when the averaged level exceeds values typical of calm work, and a critical threshold where the level of an ordinary conversation begins — and after two weeks of data the thresholds are tuned to the office's real profile, because every space sounds different. Deduplication meanwhile ensures one loud meeting does not generate a hundred separate alarms: one event is active per sensor and metric, with a code, a status and a history.
Notifications go where the team actually looks — email, SMS, web push, MS Teams, Discord or an audible in-app alarm — and quiet hours and escalation policies sort out the rest: night-time exceedances in an empty office wake nobody, while an unacknowledged alert during working hours goes one level up.
From data to a quieter office
Alerts catch incidents, but the real value of noise monitoring shows in the trends. After a month of data you see things no survey will reveal: that the quiet zone is quiet only until 10:00, that the loudest spot on the floor is not the meeting zone at all but the corridor by the small meeting room, or that noise reliably rises on Wednesdays because that is when the sales team has its calling day.

With material like this, the conversation about changes stops being an exchange of opinions. Sensor comparison widgets put zones side by side, aggregates let you move smoothly from minutes to months, and PDF reports with charts are a ready attachment for the conversation with the landlord about acoustic panels, relocating the meeting zone or buying more booths — this time where the data points. Noise is worth reading together with the other comfort parameters: a stuffy and loud zone is often one and the same layout problem, as we noted when discussing office air quality. The full comfort oversight scenario at building scale is covered in our solution for buildings and offices.
The start is smaller than it seems: four sensors, thresholds on Leq and a month of data are enough to make the first office-acoustics decisions based on numbers. The free plan includes 10 sensors and a year of measurement history — check the pricing page or book a short demo and we will show you the noise map and the alerts on live data.



