Skip to main content
GUIDES13 min read

Coworking Acoustics — Why Phone Booths Are a Symptom, Not a Solution | AcousPlan

Phone booths treat 2% of floor area while 98% stays noisy. ISO 3382-3 D2,S analysis shows ceiling treatment + masking outperforms booths 3:1 on cost.

AcousPlan Editorial · March 18, 2026

The phone booth has become the defining furniture piece of the coworking era. Walk into any WeWork, Industrious, or independent coworking space built after 2018 and you will find them lined up against the walls — glass pods, fabric-wrapped boxes, or carpeted capsules priced between $8,000 and $20,000 each. Operators display them in marketing photos as evidence of acoustic thoughtfulness.

They are evidence of the opposite. Every phone booth in a coworking space represents an acoustic design failure. The room needed those decibels absorbed. Instead, they were allowed to reach members' ears, destroy concentration, and force private conversations into a box that should never have been necessary.

This is not a minor aesthetic critique. ISO 3382-3:2012 gives us precise, measurable parameters for open-plan office acoustic performance. When you measure coworking spaces against those parameters, almost none of them pass. The phone booths are the retrofit response to a problem that acoustics should have prevented at design stage.

What ISO 3382-3:2012 Actually Measures

Most designers who claim to care about office acoustics can name RT60. Fewer have read ISO 3382-3, which was written specifically for open-plan environments and describes entirely different phenomena.

The standard defines four key measurement parameters:

D2,S (dB per distance doubling): The spatial decay rate of A-weighted speech levels across the floor. Higher values mean sound attenuates faster as you move away from a talker. Good open-plan design achieves D2,S ≥ 7 dB. Most coworking spaces measure 4–6 dB.

Lp,A,S,4m: The A-weighted SPL of a calibrated talker measured 4 m away, in dB(A). ISO 3382-3 suggests targets of ≤ 50 dB(A) for good privacy. Typical coworking floors: 53–62 dB(A).

rD (distraction distance, m): The distance at which talker SPL drops to 44 dB(A) — the threshold at which speech becomes cognitively distracting per the standard. Below 5 m is good. Most coworking spaces: 10–18 m.

rP (privacy distance, m): The distance at which SPL drops to 35 dB(A), below which conversation cannot be understood. Below 8 m is reasonable. Most coworking spaces: 20 m or effectively infinite.

What those last two numbers mean in practice: in a typical coworking space, someone making a phone call at one end of the floor is distracting everyone within 10–18 m and has zero speech privacy up to 20 m. The floor plan may be 25 m long. You have built a room where privacy physically cannot exist without enclosures.

The phone booths are not a luxury amenity. They are structural acoustic defects expressed in furniture.

How Coworking Spaces Get Built Wrong

The failure pattern is consistent across operators of all price points:

Specification step 1: Architect selects open ceiling with exposed concrete soffit for aesthetic reasons ("industrial chic," "visible services").

Concrete soffit α₅₀₀ ≈ 0.02. Zero useful absorption. Any absorption calculation that credits ceiling treatment now has nothing to work with.

Specification step 2: Flooring is polished concrete or engineered timber. α₅₀₀ ≈ 0.03–0.05.

Specification step 3: Glass partitions for natural light penetration. α₅₀₀ ≈ 0.03.

Specification step 4: A scatter of acoustic panels on two walls, typically 50mm fabric-wrapped boards specified by the interior designer for color, not performance. Perhaps 20% wall coverage.

Now run the ISO 3382-3 spatial decay calculation for a 500 m² floor (25 m × 20 m, 3.5 m floor-to-ceiling, open soffit):

  • Volume V = 3,500 m³
  • Room constant R = Sa/(1 − ā)
  • Total surface ~1,600 m²
  • Average absorption coefficient ā ≈ 0.08 (mostly hard surfaces + modest wall panels)
  • R ≈ 1,600 × 0.08 / 0.92 ≈ 139 m²
D2,S with this room constant: approximately 4.8 dB per distance doubling.

With a talker generating 65 dB(A) at 1 m (normal conversational level):

Distance from TalkerSPL dB(A)Status
1 m65Direct conversation
2 m60.2Clearly audible
4 m55.4Distracting
6 m52.5Still distracting
rD = 44 dB threshold~14 mFinally distraction-free
rP = 35 dB threshold~27 mPrivacy — beyond floor length

This is a live scenario in thousands of coworking spaces. Everyone on the floor can hear every conversation. Privacy is geometrically impossible without enclosure.

What Acoustic Design Actually Looks Like

The corrective specification is not complicated. It requires treating the sources of poor performance rather than building boxes around the symptom.

Ceiling Treatment: Where the Money Should Go

In an open-plan floor, the ceiling is the primary acoustic surface. It receives the most reflected energy from any conversation, and treating it increases D2,S by reducing the reverberant field's contribution to distant SPL measurements.

Target: NRC ≥ 0.85 across at least 70% of ceiling area, or equivalent sabin count.

For a 500 m² ceiling with full coverage at NRC 0.90:

  • Absorption at 500 Hz: 500 × 0.90 = 450 sabins from ceiling alone
  • Previous total absorption (example above): ~128 sabins
  • New total: ~578 sabins
  • New room constant R ≈ 1,600 × (128+450) / (1600 × ā_new) — let's work through it properly:
New ā = 578 / 1,600 ≈ 0.36. Room constant R = 1,600 × 0.36 / 0.64 ≈ 900 m².

D2,S with R = 900 m²: approximately 8.2 dB per distance doubling.

DistanceSPL dB(A)Status
1 m65Conversation
2 m56.8Clearly audible
4 m48.6Borderline
rD = 44 dB threshold~5.8 mDistraction-free ✓
rP = 35 dB threshold~12 mPrivacy zone

One intervention — ceiling treatment — moved rD from 14 m to 5.8 m and rP from 27 m to 12 m. The 25 m floor now has meaningful speech privacy zones. You have not built a phone booth. You have built a room that functions acoustically.

Furniture as Acoustic Infrastructure

Desk screens are not decorative. A 1.5 m high fabric-wrapped screen with NRC 0.75 at 1000 Hz places a scattering/absorbing element in the near field of every talker and listener. At 1 m distance, a screen at face height reduces SPL by 10–15 dB in the direct field — the equivalent of moving twice as far away.

The coworking specification should treat high-NRC desk screens as acoustic infrastructure, not furniture. Specify:

  • Minimum screen height: 1.4 m above floor
  • Minimum NRC: 0.70
  • Coverage: every hot desk, not just assigned stations
Soft seating — armchairs with high backs, upholstered banquettes — provides significant absorption. A 900mm × 900mm armchair with high-density foam upholstery contributes roughly 1.2 sabins at 500 Hz. Twenty such chairs distributed across a floor add 24 sabins — comparable to a 40 m² area of acoustic ceiling panel.

Specify furniture for performance. The NRC data exists for most commercial seating. Use it.

Sound Masking: The Third Leg

Sound masking is not a substitute for absorption — it is the third leg of the stool. Absorption reduces reverberant level and improves spatial decay. Masking raises the ambient noise floor uniformly, which reduces the signal-to-noise ratio of any individual conversation.

Per ASTM E1130-08 (the standard for sound masking performance), a properly installed masking system targets:

  • Spectrum: shaped to approximate speech masking bands (typically a modified AC noise curve)
  • Level: 45–48 dB(A) in focus zones, 42–45 dB(A) in quieter collaboration areas
  • Uniformity: ±2 dB variation across the floor
The masking system should be designed after the acoustic treatment is installed, not instead of it. Masking without absorption addresses the perception of privacy without actually improving speech decay. The result is a room that sounds uniformly louder, not one that provides genuine acoustic zoning.

Combined effect of absorption + masking: The reverberant SPL drops 8–10 dB. The masking floor rises 5–8 dB above ambient. The effective intelligibility distance drops significantly. Privacy is achievable without enclosures.

The Phone Booth Economics

Let us be direct about the money. A typical coworking space installs 4–8 phone booths:

Booth typeUnit cost6 booths
Basic fabric pod (2-person)$8,000$48,000
Glazed phone box (1-person)$12,000$72,000
Premium glazed office pod$20,000$120,000

Compare that to a full acoustic specification for a 500 m² floor:

InterventionAreaUnit costTotal
NRC 0.90 ceiling panels + installation500 m²$60/m²$30,000
High-NRC desk screens (40 units)$400/unit$16,000
Sound masking system (installed)500 m²$25/m²$12,500
Acoustic wall panels (80 m²)80 m²$80/m²$6,400
Total~$65,000

The comprehensive acoustic design costs roughly the same as six glazed phone booths — and delivers a floor that doesn't need phone booths. Members can work anywhere without being disrupted. Private calls can be made at the desk with confidence. The experience is categorically better, not marginally better.

The calculus only goes wrong when the acoustic design is specified after the exposed concrete ceiling is signed off. At that point, the ceiling treatment requires a grid suspension system or direct adhesive mounting, the cost doubles, and the project manager approves the phone booths instead.

The time to specify acoustics is before the ceiling finish is chosen. That is when the decision costs almost nothing.

Measuring What You Have Built

If you are inheriting a coworking design brief or doing a retrofit assessment, ISO 3382-3 measurements take approximately half a day on a floor and produce the four parameters that tell you exactly what you are dealing with.

The measurement protocol (per ISO 3382-3 §6.2):

  1. Set up a calibrated omnidirectional loudspeaker at a representative workstation position
  2. Play a calibrated pink noise or swept sine signal at 65 dB(A) at 1 m (simulating normal speech)
  3. Measure A-weighted SPL at 2, 4, 6, 8 m along two perpendicular transects
  4. Calculate D2,S by regression on the SPL-vs-log-distance data
  5. Read rD and rP directly from the measurement transect where SPL crosses 44 and 35 dB respectively
Any acoustic consultant with basic equipment can produce this data. It is far more actionable than a generic statement about RT60. If your brief says "acoustics: phone booths to be provided," push back and ask for ISO 3382-3 measurement criteria instead. Specify a maximum rD of 5 m as a performance target. That number is meaningful, measurable, and achievable.

Zoning Strategy: Different Parts of the Floor Need Different Treatment

Not all coworking is equal. A well-designed floor differentiates between:

Focus zones: Targeted rD ≤ 4 m, no music, masking level 45 dB(A). Treatment: full ceiling absorption + high screens + masking emitters directly overhead.

Collaboration zones: rD ≤ 8 m acceptable, light background music at 50 dB(A) allowed. Treatment: ceiling absorption + lower screens or open benching, masking at lower level.

Social/breakout zones: No privacy requirements, RT60 ≤ 1.0 s for intelligibility. Treatment: absorptive soft furnishings to prevent intelligibility from being completely destroyed.

Each zone has different treatment requirements and different ISO 3382-3 targets. The zoning plan should be on the acoustic drawing set alongside the reflected ceiling plan, with material specifications attached.

Practical Use of AcousPlan for Coworking Design

Before specifying a single panel, model the floor in the speech privacy calculator. Input room dimensions, existing surface finishes, and target rD. The tool calculates current speech decay parameters and the treatment delta required to hit your target.

For coworking spaces, the most useful output is the comparison between current rD and target rD. If the gap is larger than 5 m, ceiling treatment alone will not close it — you need the combination of ceiling absorption, furniture screening, and masking. The calculator quantifies how much of each is required.

The Lease Specification Problem

There is a structural reason coworking spaces are built badly: most operators are tenants, not building owners. They lease raw commercial floor space and fit it out under a landlord licence to alter. The acoustic performance of the base build — specifically the ceiling finish — is locked in before the operator's fit-out team arrives.

When a landlord spec includes exposed concrete soffit as a cost-saving measure or aesthetic choice, the coworking operator inherits an untreatable ceiling without owning the structure to alter it. Suspended acoustic grid systems require drilling or anchoring into the slab — work that typically requires structural engineer sign-off and landlord approval with lead times of months.

By the time the fit-out team realises the problem (usually when they do a site visit after launch and discover members complaining immediately), the lease is signed, the fit-out budget is spent, and the phone booths arrive as the only practical option within commercial constraints.

The fix is earlier engagement. For operators negotiating new leases, the acoustic performance of the base build ceiling should be a non-negotiable specification item. A short clause requiring the landlord to deliver the space with ceiling absorption NRC ≥ 0.80, with suspension infrastructure for supplementary baffles, costs the operator nothing in rent but prevents the phone booth cascade entirely.

For the acoustic consultant retained by a coworking operator at brief stage — before lease execution — the highest-value deliverable is a one-page specification clause for the base-build ceiling. That single page has more impact on member experience than any amount of post-occupancy treatment.

What Operators Should Ask Architects

If you are the architect briefed on a coworking fit-out, expect to be asked questions you may not currently answer with sufficient precision. Operators who have been through the phone booth experience once start asking:

  • What will the rD (distraction distance per ISO 3382-3) be on the focus floor?
  • What is the background noise level (NR curve) from the HVAC system we are inheriting?
  • If we specify this ceiling treatment, what STI does the collaboration zone achieve?
  • What is the acoustic zoning boundary — where exactly does the focus zone end and the social zone begin?
These are correct questions. They have numerical answers. If your acoustic specification cannot provide them, you are not specifying acoustics; you are specifying materials and hoping.

The coworking sector is building hundreds of spaces annually, most of them with the same acoustic failures. The market for genuine acoustic specification in this sector is large, underserved, and increasingly demanded by operators who have watched their phone booth spend escalate. The argument is straightforward: measure the problem, specify the fix, deliver a room that functions. Leave the booths for genuine privacy requirements, not for compensating for bad room acoustics.

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