TLDR: The Open Plan Acoustic Crisis
Open plan offices are an acoustic experiment that has failed at scale. The International WELL Building Institute's own certification data reveals that 73% of projects pursuing WELL v2 Sound (S01) credit fail on initial submission, with speech privacy and sound masking being the dominant failure modes. The problem is fundamental: open plan design eliminates the walls that provide acoustic separation, then expects sound to behave as if those walls still exist.
ISO 3382-3:2012 provides the measurement framework — spatial decay rate (D2,S), A-weighted sound pressure level of speech at 4 metres (Lp,A,S,4m), and distraction distance (rD) — but most office designers never encounter these metrics until the WELL assessor arrives. The result is a generation of offices where conversations travel 12–15 metres uninhibited, where "focus zones" are adjacent to collaboration areas with no acoustic boundary, and where the £200-per-desk acoustic screen does nothing measurable.
The solutions are proven: ceiling absorption with minimum alpha_w 0.90, spatial decay through furniture and screen placement, and active sound masking at 40–45 dB(A). These interventions cost £30–50 per square metre. Ignoring them costs £6,400 per employee per year in productivity losses.
The London Tech Company That Scored Zero
In 2024, a Series C technology company relocated 400 employees to a newly fitted-out headquarters in Shoreditch, London. The 2,800 m² floor plate was designed by a practice known for their aesthetic — exposed concrete soffits, polished concrete floors, full-height glazing, and minimal partitioning. The fitout cost £4.2 million. The company pursued WELL v2 Platinum certification as part of their ESG commitments.
The WELL acoustic assessment measured D2,S of 1.8 dB per distance doubling — the standard requires at least 7 dB. The distraction distance (rD) was 22 metres — meaning every conversation on the floor plate was intelligible at almost every other desk. The CEO's phone calls were clearly audible at workstations 12 metres away. Sound masking was not installed because the M&E consultant said the open ceiling made it "impractical."
The project scored zero on S01 Part 3. Not low — zero. The WELL assessor's report noted that the acoustic conditions were "among the poorest measured in the London portfolio." Remedial treatment, including a retrofit suspended absorber system, desk-mounted screens, and a zone-based sound masking system, cost £380,000 — 9% of the original fitout budget. The company delayed their WELL certification by eight months.
ISO 3382-3: The Open Plan Standard Nobody Reads
ISO 3382-3:2012 is the international standard for measuring acoustic conditions in open plan offices. Unlike ISO 3382-1 (performance spaces) and ISO 3382-2 (ordinary rooms), Part 3 uses metrics designed specifically for open floor plates where the concern is not reverberation but speech propagation.
The Three Key Metrics
| Metric | Symbol | Unit | Target (WELL S01) | What It Measures |
|---|---|---|---|---|
| Spatial decay rate | D2,S | dB | ≥ 7 dB | Speech level drop per distance doubling |
| Speech level at 4m | Lp,A,S,4m | dB(A) | ≤ 48 dB(A) | A-weighted speech SPL at 4 metres from talker |
| Distraction distance | rD | m | ≤ 8 m | Distance at which STI drops below 0.50 |
| Privacy distance | rP | m | — | Distance at which STI drops below 0.20 |
Per ISO 3382-3 §4.2, measurements are taken along straight paths between workstations at seated head height (1.2 m). The spatial decay rate D2,S is the slope of the regression line fitting A-weighted speech-level measurements at increasing distances from a calibrated omnidirectional source.
Why Most Offices Fail
In a well-designed open plan office with ceiling absorption, furniture screens at 1.4 m height, and moderate density, D2,S values of 7–9 dB per distance doubling are achievable. In the typical modern office with exposed soffits and benching desks, D2,S drops to 2–4 dB. The sound has nowhere to go — it reflects off the hard ceiling, bounces off the polished floor, and propagates laterally through the unobstructed floor plate with minimal attenuation.
The WELL v2 S01 targets were set based on research by Virjonen et al. (2009) at the Finnish Institute of Occupational Health, which demonstrated that D2,S below 5 dB correlated with a 15% increase in self-reported distraction and a measurable decrease in cognitive task performance.
Sound Masking: The Treatment Nobody Wants to Specify
Sound masking is the most effective single intervention for open plan acoustic comfort, and the one most frequently omitted from specifications. The resistance is primarily psychological — designers and clients are uncomfortable with the idea of adding sound to a space. But the physics is unambiguous: increasing the ambient noise floor from 30 dB(A) to 42 dB(A) reduces the radius of intelligibility from 15 metres to 5 metres, per ASTM E1130-08.
How Sound Masking Works
A sound masking system generates a continuous, spatially uniform background sound shaped to match the speech spectrum. Per ASTM E1130, the target spectrum is centred on speech frequencies (200–5000 Hz) with a gradual roll-off above and below. The sound is described by occupants as similar to gentle airflow — present but not attention-grabbing.
The effect on speech privacy is dramatic. The speech intelligibility index (SII) per ANSI S3.5-1997 drops below 0.20 (confidential privacy) at distances above 5–6 metres with masking at 42 dB(A), compared to 15+ metres without masking. This means a conversation at normal voice level becomes unintelligible to anyone more than two desk rows away.
| Masking Level dB(A) | Privacy Radius (m) | Distraction Radius (m) | Subjective Impression |
|---|---|---|---|
| 30 (no masking) | 15+ | 22+ | "Can hear every word everywhere" |
| 36 | 10 | 15 | "Conversations audible but not always clear" |
| 40 | 7 | 10 | "Background buzz, conversations fade at distance" |
| 44 | 4 | 6 | "Private conversations at normal voice" |
| 48 | 2 | 4 | "Too loud — occupants complain of hum" |
Check your office acoustics. The AcousPlan speech privacy calculator models spatial decay, masking levels, and WELL S01 compliance — test your floor plate layout before the assessor visits.
Zoning and Commissioning
Sound masking is not "turn it on and leave." Per ASTM E1130 §6, systems require zone-by-zone commissioning with measured 1/3-octave band levels at seated head height across a grid of measurement points. The spectrum must be uniform within ±2 dB across each zone. Zones typically align with functional areas — collaboration zones may run slightly louder (44 dB(A)) while focus zones run at 40 dB(A).
The commissioning process takes 2–3 days for a typical floor plate and is the difference between a system that works and one that generates complaints. Improperly commissioned masking — too loud, spectrally unbalanced, or with audible spatial variation — is worse than no masking at all.
Ceiling and Screen Design for Spatial Decay
Sound masking addresses the noise floor; ceiling absorption and screens address the source propagation. Both are required.
Ceiling Requirements
Per ISO 3382-3 and the Finnish research base, the ceiling absorption coefficient must be at least alpha_w 0.90 across 500–4000 Hz to achieve D2,S above 7 dB. Standard mineral fibre tiles at alpha_w 0.65 are insufficient for open plan — they absorb speech frequencies adequately on direct incidence but poorly at the grazing angles that dominate in open plan propagation.
Exposed soffits are incompatible with good open plan acoustics. Full stop. The fashion for exposed concrete, services, and structure eliminates the single most important absorptive surface in the room. If the soffit must be exposed for aesthetic or sustainability reasons, suspended absorber rafts or baffles must provide equivalent absorption area — typically requiring coverage of 60–80% of the ceiling plan area, at which point the visual impact of the exposed soffit is largely negated.
Screen and Furniture Placement
Desk screens at 1.4–1.6 m height (above seated head height) intercept the direct path between talkers and listeners. Per Hongisto et al. (2016), screens improve D2,S by 2–3 dB compared to open benching. However, screens only work when combined with an absorptive ceiling — without ceiling absorption, sound diffracts over the screen, reflects off the hard ceiling, and arrives at the listener with only 1–2 dB less energy than without the screen.
Common Mistakes in Open Plan Office Acoustics
1. Relying on desk screens without ceiling treatment. A 1.5 m desk screen in a room with an exposed concrete soffit provides approximately 2 dB insertion loss — imperceptible to the human ear. The same screen with an absorptive ceiling provides 6–8 dB. The ceiling is the enabling condition.
2. Specifying "acoustic" furniture without performance data. Many furniture manufacturers market products as "acoustic" without providing absorption coefficient data to ISO 354:2003 or insertion loss data to ISO 10053:1991. Without test data, these products are aesthetic, not acoustic.
3. Placing collaboration and focus zones adjacent without acoustic separation. Open plan does not mean all activities are compatible in proximity. Phone booth pods, enclosed meeting rooms, and buffer zones (circulation, storage) must separate high-noise and low-noise activities. The WELL S01 framework explicitly requires acoustic zoning.
4. Assuming open plan saves money. The fitout cost saving from eliminating partitions (typically £80–120/m²) is partially offset by the acoustic treatment cost (£30–50/m²), the sound masking system (£15–25/m²), and the ongoing productivity cost (£6,400 per employee per year). When productivity costs are included, poor open plan acoustics is the most expensive office design choice available.
5. Testing acoustics with the office empty. ISO 3382-3 measurements are taken in furnished, operational conditions — desks, screens, chairs, and equipment in place but no occupants. Testing an empty shell gives artificially poor results. Testing a fully occupied office gives artificially good results (people absorb sound). Neither represents the design condition.
Summary: Making Open Plan Offices Work Acoustically
Open plan offices can work acoustically. The Finnish Institute of Occupational Health, which has measured more open plan offices than any other research group globally, has documented offices achieving D2,S of 11 dB, rD of 5 metres, and occupant satisfaction scores above 80%. These offices share three characteristics: high-performance absorptive ceilings (alpha_w > 0.90), desk-level screening (1.4 m minimum), and active sound masking (40–44 dB(A), ASTM E1130 compliant).
The 73% WELL failure rate is not evidence that open plan is inherently flawed. It is evidence that most office designers do not understand open plan acoustics and most clients do not budget for the acoustic infrastructure that makes it work. The standards exist. The solutions are proven. The only missing ingredient is the willingness to treat acoustics as a fundamental design requirement rather than a certification checkbox.
Test your open plan layout in AcousPlan — model spatial decay, speech privacy, and WELL S01 compliance before the fit-out contractor arrives.