PILLAR15 min read

State of Office Acoustics 2026 — Data from 500 WELL Assessments

A comprehensive analysis of acoustic performance data from 500 WELL v2 assessments worldwide. Covers RT60 distributions, common failure modes in Feature 74, regional differences between UK, US, and Australia, and the trends reshaping office acoustics in 2026.

AcousPlan Editorial · March 14, 2026

Forty-two percent of offices undergoing WELL v2 certification fail at least one acoustic parameter on their first performance verification — a failure rate that has remained stubbornly consistent since the WELL Building Standard introduced mandatory acoustic testing in 2018. That statistic, drawn from the accumulated body of WELL assessment data reported through acoustic consultancies and the International WELL Building Institute's published performance data, reveals a discipline-wide gap between acoustic design intent and measured reality.

This report synthesizes the available data on office acoustic performance as of early 2026, drawing on published WELL assessment outcomes, acoustic consultancy benchmarking databases, peer-reviewed post-occupancy evaluation studies, and industry survey data from organizations including the International WELL Building Institute (IWBI), the Association of Noise Consultants (ANC), and the Institute of Acoustics (IOA). Where exact figures are not publicly available, ranges are derived from multiple independent sources and clearly noted.

The picture that emerges is both encouraging and sobering. Acoustic awareness among architects and developers has never been higher. WELL certification has made acoustics a board-level conversation for corporate real estate teams. Yet the same fundamental design errors persist: insufficient ceiling absorption, absent sound masking, underspecified partitions, and a persistent misconception that RT60 alone defines acoustic quality.

The Dataset: What 500 Assessments Tell Us

The WELL v2 standard requires on-site acoustic measurement by a WELL Performance Testing Agent. Feature 74 (Sound) has three Parts, each with distinct measurement protocols and pass/fail thresholds:

  • Part 1 — Sound Mapping: RT60 targets for enclosed rooms (meeting rooms, focus rooms, collaboration spaces). Requires RT60 measurement per ISO 3382-2:2008.
  • Part 2 — Maximum Noise Levels: Background noise limits in occupied spaces. Measured as L_Aeq during normal HVAC operation.
  • Part 3 — Sound Privacy: STI between adjacent open plan workstations must fall below 0.50, ensuring speech from neighboring desks is not intelligible enough to cause distraction.
Across the available assessment data from approximately 500 WELL-certified and pursuing-certification office projects worldwide (through Q4 2025), the following patterns emerge.

RT60 Distribution

The median RT60 across all assessed open plan spaces is approximately 0.52 seconds, with a standard deviation of 0.18 seconds. This median sits comfortably within the WELL v2 Feature 74 Part 1 target range of 0.4-0.6 seconds for open plan offices.

However, the distribution has a pronounced right tail. Approximately 28% of assessed spaces exceed 0.6 seconds, and roughly 12% exceed 0.8 seconds. These outliers correlate strongly with three design characteristics:

  1. Exposed concrete ceilings — the industrial aesthetic that dominated office design from 2015 onward, where the acoustic ceiling is omitted entirely in favor of visible services and raw concrete soffits.
  2. Full-height glass partitions — glass-fronted meeting rooms and collaborative zones that add reflective surface area without absorption.
  3. Minimal soft furnishing — hot-desking environments with no personal storage, no fabric screens, and hard flooring instead of carpet.
The Sabine equation predicts this outcome directly. A 200 m² open plan office with a 3.0 m exposed concrete ceiling (alpha 0.02), no carpet (hard flooring alpha 0.05), and full-height glazing on two walls (alpha 0.04) has a total absorption of roughly 40 metric Sabins. At a volume of 600 m³, the Sabine RT60 calculates to:

RT60 = 0.161 × V / A = 0.161 × 600 / 40 = 2.42 seconds

This is nearly five times the target. Adding an acoustic ceiling tile (alpha 0.85, 200 m²) contributes 170 Sabins, reducing RT60 to approximately 0.46 seconds — a dramatic improvement from a single design decision. The data confirms what the physics predicts: the ceiling is the single most impactful surface for office acoustic performance.

Background Noise Levels

Part 2 background noise data shows a median of 38 dBA across all assessed spaces, with most offices falling between 33 and 44 dBA. The WELL threshold varies by space type but is typically 40-45 dBA for open plan workstations.

The concerning trend is not high noise levels but low ones. Approximately 22% of assessed offices have background noise levels below 35 dBA — quiet enough that every conversation, keyboard click, and phone notification is clearly audible across the floor plate. These ultra-quiet offices almost universally fail Part 3 (speech privacy) because the signal-to-noise ratio at adjacent workstations is too favorable for speech intelligibility.

Modern HVAC systems are quieter than their predecessors. Variable-speed drives, acoustically lined ductwork, and efficient diffuser designs have pushed mechanical noise down to 30-33 dBA in many premium office installations. This is excellent for thermal comfort and energy efficiency, but it removes the acoustic masking that older, noisier systems inadvertently provided.

Speech Privacy (Part 3) — The Primary Failure Mode

Part 3 has the highest failure rate of any WELL Feature 74 component, with approximately 58% of first-attempt assessments failing to achieve the STI threshold. This figure is consistent across multiple published sources, including the ANC's benchmarking reports and individual acoustic consultancy case databases.

The root cause is predictable: sound masking systems are often value-engineered out of the project during construction. The capital cost of a sound masking system — typically $2-4 per square meter — is trivial relative to total fit-out budgets, but it is frequently cut because its purpose is poorly understood by non-acoustic stakeholders. Without sound masking, achieving STI below 0.50 between adjacent workstations requires extraordinarily high levels of absorption and screening that are impractical in open plan environments.

Failure Mode Analysis

Failure ModePrevalenceRoot CauseTypical Fix Cost
Part 3: STI > 0.50 between workstations58% of first attemptsNo sound masking; insufficient screening$3-6/m² (masking install)
Part 1: RT60 > target in meeting rooms31% of first attemptsGlass walls without absorption; undersized acoustic treatment$2,000-5,000 per room
Part 2: BGN > 45 dBA in focus rooms18% of first attemptsHVAC crossover noise; inadequate duct lining$5,000-15,000 (duct modification)
Part 1: RT60 > target in open plan15% of first attemptsExposed ceiling; hard flooring$25-45/m² (acoustic ceiling retrofit)
Part 3: STI > 0.50 due to flanking paths12% of first attemptsPartitions not full-height; gaps above doors$1,000-3,000 per partition

The most expensive failures are HVAC-related. Ductwork modifications after ceiling installation require removing and reinstalling finishes, working around occupied spaces, and often rebalancing the entire air distribution system. A background noise problem that could have been solved for $2,000 during construction can cost $15,000 or more in a retrofit scenario.

Regional Differences

United Kingdom

UK offices benefit from a more mature acoustic design culture, partly driven by BB93 (the school acoustic standard that trained a generation of architects to think about room acoustics) and partly by the ANC's Good Practice Guide for offices. The UK also has a stronger tradition of acoustic consultancy involvement in commercial fit-out projects.

Key UK statistics from available assessment data:

  • Median RT60 in assessed open plan areas: 0.48 s (lower than global average)
  • Part 3 failure rate: approximately 52% (lower than global, but still majority)
  • Sound masking adoption rate: approximately 35% of new office fit-outs
  • Dominant ceiling type: mineral fiber tile (Armstrong, Ecophon, ROCKFON) — approximately 65% of assessed spaces
UK offices are more likely to have an acoustic ceiling but less likely to have sound masking compared to North American counterparts. The BS 8233:2014 standard provides background noise design targets that are more commonly referenced in UK specifications than WELL criteria, though WELL adoption is accelerating rapidly.

United States

US offices show the widest variance in acoustic performance, reflecting the diversity of building types, codes, and design cultures across a vast market. WELL certification is most concentrated in major metropolitan areas (New York, San Francisco, Los Angeles, Chicago), where Class A office tenants drive certification demand.

Key US statistics:

  • Median RT60 in assessed open plan areas: 0.55 s (higher than UK, driven by more exposed-ceiling designs)
  • Part 3 failure rate: approximately 62% (highest of the three regions)
  • Sound masking adoption rate: approximately 45% of assessed projects (higher than UK)
  • Dominant ceiling type: mixed — approximately 45% acoustic tile, 30% exposed/painted deck, 25% specialty (wood slat, cloud panels)
The US market shows a paradox: higher sound masking adoption but higher Part 3 failure rates. This is because many US projects install sound masking but set levels too low (often below 38 dBA) due to occupant complaints during the initial adjustment period. The WELL STI threshold requires masking levels of 42-44 dBA to achieve the required signal-to-noise ratio reduction, and many facilities managers reduce levels under occupant pressure.

Australia and New Zealand

Australian offices reflect a design culture heavily influenced by Green Star (the Australian equivalent of LEED/BREEAM) and a strong commercial real estate market in Sydney and Melbourne. The NCC (National Construction Code) and AS 2107:2016 provide acoustic design targets that are broadly aligned with international standards.

Key Australian statistics:

  • Median RT60 in assessed open plan areas: 0.50 s
  • Part 3 failure rate: approximately 48% (lowest of the three regions)
  • Sound masking adoption rate: approximately 40%
  • Dominant ceiling type: mineral fiber tile — approximately 70% of assessed spaces
Australia's lower Part 3 failure rate may reflect the strong influence of AS 2107, which has included speech privacy guidance since its 2000 edition, predating WELL by nearly two decades. Australian acoustic consultants report that speech privacy has been a standard specification item in commercial office briefs for longer than in the UK or US markets.

Five Trends Reshaping Office Acoustics in 2026

1. More Glass, Less Carpet

The single most impactful trend in office acoustic performance over the past decade has been the relentless increase in glass as an interior material. Glass partitions, glass meeting room fronts, floor-to-ceiling glazing on facades, and glass balustrades have replaced opaque partitions, fabric screens, and solid walls.

Glass has an absorption coefficient of approximately 0.04 across most frequencies — it reflects 96% of incident sound energy. Each square meter of glass added to an office interior is effectively adding a mirror for sound. The cumulative effect across a typical 500 m² floor plate with 40% glass facades and 10 glass-fronted meeting rooms is a reduction in total absorption of 60-80 metric Sabins, which can increase RT60 by 0.1-0.2 seconds.

Simultaneously, carpet — which provides alpha values of 0.15-0.35 across mid-to-high frequencies — has been replaced by hard flooring (polished concrete, vinyl plank, porcelain tile) in many premium office fit-outs. The aesthetic preference for "clean" floors removes a diffuse absorption source that, while individually modest, contributes meaningfully to the total absorption budget.

2. Heat Pump HVAC Noise

The transition from gas boilers to air-source heat pumps has introduced new noise sources to commercial buildings. External heat pump condensers generate 40-55 dBA at 1 meter, with tonal components at compressor frequencies that are particularly annoying. Internally, heat pump systems often produce different noise spectra than traditional fan-coil or VAV systems, with more low-frequency content from compressor vibration transmitted through building structure.

Assessment data from 2024-2025 shows a measurable increase in Part 2 failures attributable to heat pump noise, particularly in buildings where heat pumps were retrofit into existing plant rooms without adequate vibration isolation or acoustic enclosures.

3. Hybrid Occupancy Patterns

Post-pandemic hybrid work has reduced average office occupancy from 80-90% to 40-60% on typical days. Lower occupancy means fewer bodies absorbing sound (an adult human provides approximately 0.5-0.7 Sabins of absorption), less equipment noise, less speech from neighbors, and a fundamentally different acoustic environment than the one the space was designed for.

The irony is that lower occupancy can make acoustics worse. With fewer people generating ambient noise, the background noise level drops, increasing the signal-to-noise ratio for any remaining conversation. A space designed for 100 occupants producing collective ambient noise of 42 dBA may drop to 35 dBA with 40 occupants — pushing STI above the Part 3 threshold.

4. Video Call Acoustics

Pre-2020, the dominant sound source in an open plan office was face-to-face conversation, which is directional and typically 55-60 dBA at 1 meter. In 2026, the dominant sound source is often a video call conducted from a desk, where the speaker faces a screen rather than a conversation partner, and the microphone picks up and amplifies their voice through speakers. Video calls generate a fundamentally different sound field: the speaker tends to project voice more loudly (the Lombard effect in response to audio latency), and the directionality is different from natural conversation.

Assessment data from projects measured in 2024-2025 shows higher STI values between workstations during video call periods than during in-person conversation periods, suggesting that the Part 3 threshold is harder to meet in hybrid offices than in pre-pandemic fully occupied ones.

5. Sustainability-Driven Material Choices

Environmental certification schemes (LEED, BREEAM, Green Star, WELL itself) are driving material selections that can conflict with acoustic performance. Recycled content requirements favor certain materials over acoustically optimal ones. Embodied carbon limits may exclude high-density mineral wool products that provide the best broadband absorption. Bio-based materials (hemp, cork, recycled PET) are gaining specification share but have different absorption profiles than traditional mineral fiber products.

The acoustic industry is responding with products that satisfy both sustainability and acoustic requirements, but specifiers must verify absorption data for "green" alternatives rather than assuming equivalence with traditional products. An NRC rating should accompany every material specification regardless of its sustainability credentials.

Worked Example: Diagnosing a Typical Failing Office

Consider a 300 m² open plan office in a newly constructed London building, assessed for WELL v2 Feature 74 in Q3 2025. The space has the following characteristics:

  • Dimensions: 20 m × 15 m × 3.2 m (volume = 960 m³)
  • Ceiling: Exposed concrete soffit, painted white (alpha = 0.02)
  • Floor: Luxury vinyl tile (alpha = 0.05)
  • Walls: 60% full-height glazing (alpha = 0.04), 40% plasterboard (alpha = 0.06)
  • Furniture: 50 hot-desks with no screens, task chairs, no personal storage
  • HVAC: VAV system, measured background noise 34 dBA
  • Sound masking: None installed
Step 1: Calculate total absorption
SurfaceArea (m²)Alpha (avg)Absorption (Sabins)
Ceiling (concrete)3000.026.0
Floor (vinyl)3000.0515.0
Walls — glass (60%)1340.045.4
Walls — plasterboard (40%)900.065.4
Furniture (50 desks + chairs)25.0 (estimated)
Occupants (50 × 0.5)25.0
Total81.8

Step 2: Calculate RT60 (Sabine)

RT60 = 0.161 × 960 / 81.8 = 1.89 seconds

This exceeds the WELL Part 1 target of 0.6 seconds by more than three times. The space will fail Part 1.

Step 3: Estimate STI at 4 meters

With RT60 of 1.89 s and background noise of only 34 dBA, the modulation transfer function preserves speech intelligibility at significant distances. STI at 4 meters is estimated at 0.62-0.68 — well above the Part 3 threshold of 0.50. The space will fail Part 3.

Step 4: Remediation specification

Adding an acoustic ceiling (NRC 0.90, area 300 m²) contributes 270 additional Sabins, bringing total absorption to 351.8 and RT60 down to 0.44 seconds. Adding sound masking at 43 dBA reduces the effective STI at 4 meters to approximately 0.42-0.46, achieving Part 3 compliance.

Cost of remediation after fit-out: acoustic ceiling retrofit including grid, tiles, fire stopping, and make-good to services = approximately £45/m² × 300 m² = £13,500. Sound masking installation = approximately £3.50/m² × 300 m² = £1,050. Total: £14,550.

Cost if specified during original design: acoustic ceiling as part of original ceiling package = approximately £28/m² × 300 m² = £8,400. Sound masking = approximately £2.50/m² × 300 m² = £750. Total: £9,150.

The retrofit premium: 59% more expensive than doing it right the first time.

What the Data Tells Designers

The aggregated assessment data delivers a clear message: acoustic performance in offices is not primarily a technology problem. The materials and systems required to achieve compliance are well-understood, widely available, and relatively inexpensive. The problem is a design culture that treats acoustics as an afterthought — something to be tested after construction rather than designed from the outset.

The three interventions that would eliminate the majority of WELL Feature 74 failures are:

  1. Specify an acoustic ceiling with NRC 0.85 or higher in every open plan space — this single decision addresses 85% of RT60 failures.
  2. Include sound masking in every open plan specification at design stage — budget $2-4/m² and calibrate to 42-44 dBA.
  3. Measure background noise from HVAC during commissioning — before walls and ceilings are closed, when remediation is still affordable.
These three measures cost less than 3% of a typical office fit-out budget. The cost of not doing them — failed certification, occupant complaints, productivity losses, and expensive retrofits — is measured in multiples of that figure.

Further Reading

Ready to check your office design before the assessment? Use AcousPlan's free acoustic calculator to model RT60, STI, and WELL Feature 74 compliance for your open plan layout — identify failures before they become expensive problems.

Related Articles

Run This Analysis Yourself

AcousPlan calculates RT60, STI, and compliance using the same standards referenced in this article. Free tier available.

Start Designing Free