The Complaint That Costs Gym Owners Members
73% of group fitness participants in a 2022 ukactive member survey cited "can't hear the instructor" as a reason for moving to the back of the class or skipping classes entirely. The survey covered 112 UK health clubs and found that studios with measured RT60 above 1.5 seconds had member retention rates 22% lower than studios with RT60 below 1.0 seconds. The echo was not just annoying — it was directly costing gym operators revenue.
The physics is straightforward. A fitness instructor giving verbal cues during a high-energy class — "four more reps," "switch sides," "thirty seconds left" — is competing against music at 85–95 dBA, the impact noise of 20–30 participants moving simultaneously, and the reverberant tail of their own previous instruction. In a room with RT60 of 2.5 seconds, the reverberant energy from the music alone takes 2.5 seconds to decay by 60 dB. Every verbal cue is buried under a wall of overlapping reflections.
This article covers the acoustic physics of fitness studios, a worked calculation for a typical studio space, and the treatment specification that solves the problem for under £90/m².
Why Fitness Studios Are Acoustically Difficult
The Worst Combination of Factors
Fitness studios combine every acoustic challenge into a single space:
High sound levels: Music in group fitness classes typically runs at 85–95 dBA measured at the centre of the room. Instructor voice levels reach 90–100 dBA through headset microphone and PA system. Impact noise from jumping, weights hitting the floor, and equipment use adds broadband energy across all frequencies.
Hard, reflective surfaces: Modern fitness studios favour an industrial aesthetic — exposed concrete ceilings, polished concrete or rubber floors, glass partition walls, and painted blockwork. These surfaces have absorption coefficients below 0.05 at mid frequencies. The room reflects almost all sound energy back into the space.
Moderate volume with low absorption: A typical group fitness studio (80–120 m²) has enough volume to sustain a significant reverberant field but not enough volume to provide the natural air absorption that helps control reverberation in very large spaces.
Multiple simultaneous sources: Unlike a lecture room where a single speaker addresses a quiet audience, a fitness studio has 20–30 people generating impact noise, a PA system, and background ventilation — all contributing to the overall sound field simultaneously.
The Speech Intelligibility Problem
For an instructor's verbal cues to be understood, the STI at the listener position must be at least 0.45 (the "fair" threshold per IEC 60268-16:2020 §4.4). In practice, fitness environments need STI closer to 0.55–0.60 because participants are physically exerting themselves — elevated heart rate and breathing reduce auditory processing capacity, so the acoustic signal needs to be cleaner than in a calm listening environment.
In a studio with RT60 of 2.5 seconds and music at 90 dBA, the STI from the instructor's PA system to a participant at the back of the room (8 metres from the nearest speaker) is typically 0.25–0.35 — solidly in the "poor" category. The participant hears sound, recognises it as speech, but cannot reliably decode the words. They follow the instructor's visual cues instead, which works for simple movements but fails for timing cues, motivational speech, and safety instructions ("stop," "modify," "slow down").
Worked Example: 10 m x 8 m x 4.5 m Fitness Studio
The Space
A typical group fitness studio in a UK commercial gym:
- Dimensions: 10 m long x 8 m wide x 4.5 m high
- Volume: 360 m³
- Floor area: 80 m²
- Capacity: 25 participants
Existing Surface Schedule (Untreated)
| Surface | Area (m²) | Material | α at 500 Hz | α at 1 kHz | α at 125 Hz |
|---|---|---|---|---|---|
| Ceiling | 80 | Exposed concrete soffit | 0.02 | 0.02 | 0.01 |
| Floor | 80 | Rubber gym flooring (6 mm) on concrete | 0.04 | 0.04 | 0.03 |
| Long walls | 2 x (10 x 4.5) = 90 | Painted blockwork | 0.03 | 0.04 | 0.02 |
| Short walls | 2 x (8 x 4.5) = 72 | 1 glass (8 x 4.5 = 36 m²), 1 painted blockwork (36 m²) | See below | See below | See below |
| Mirrors | 20 (on one long wall) | Glass mirror on blockwork | 0.03 | 0.02 | 0.03 |
| Total surface | 342 |
Short wall breakdown:
- Glass partition wall: 36 m², α at 500 Hz = 0.04, α at 1 kHz = 0.03
- Painted blockwork wall: 36 m², α at 500 Hz = 0.03, α at 1 kHz = 0.04
RT60 Calculation — Sabine Equation
Per ISO 3382-2:2008 §A.1:
At 500 Hz:
| Surface | Area (m²) | α | A (sabins) |
|---|---|---|---|
| Concrete ceiling | 80 | 0.02 | 1.60 |
| Rubber floor | 80 | 0.04 | 3.20 |
| Long wall — blockwork | 70 | 0.03 | 2.10 |
| Long wall — mirror section | 20 | 0.03 | 0.60 |
| Short wall — glass | 36 | 0.04 | 1.44 |
| Short wall — blockwork | 36 | 0.03 | 1.08 |
| Total | 322 | 10.02 |
T60 at 500 Hz = 0.161 x 360 / 10.02 = 57.96 / 10.02 = 5.8 seconds
This extreme result reflects the near-total reflectivity of all surfaces. In practice, equipment (spin bikes, weight racks, foam rollers) and ductwork add some scattering and absorption. A more realistic furnished estimate adds approximately 5–8 sabins, yielding:
T60 at 500 Hz (furnished, unoccupied) = 0.161 x 360 / 16 = 3.6 seconds
With 25 participants (standing/exercising, approximately 0.50 sabins per person at 500 Hz):
Additional absorption = 25 x 0.50 = 12.5 sabins
T60 at 500 Hz (occupied) = 0.161 x 360 / (16 + 12.5) = 57.96 / 28.5 = 2.0 seconds
Even fully occupied during a class, the RT60 is 2.0 seconds — double the target of 1.0 seconds.
At 1 kHz:
Total absorption (furnished, occupied) increases slightly due to higher absorption of rubber flooring and occupants at 1 kHz. Estimated A = 33 sabins.
T60 at 1 kHz (occupied) = 0.161 x 360 / 33 = 1.76 seconds
Still well above the target. The problem persists across the speech-critical frequency range.
Treatment Design: Achieving RT60 = 0.8 Seconds
Target: T60 = 0.8 seconds at 500 Hz (occupied)
Required total absorption: A = 0.161 x 360 / 0.8 = 57.96 / 0.8 = 72.5 sabins
Current absorption (occupied): 28.5 sabins
Additional absorption needed: 72.5 - 28.5 = 44.0 sabins
Treatment Option: Suspended Acoustic Baffles
Specification:
- Baffle dimensions: 1200 mm high x 600 mm wide x 50 mm thick
- Core: mineral wool, density 60 kg/m³, wrapped in Class 0 acoustic fabric
- Mounting: suspended vertically from ceiling soffit on stainless steel wire at 600 mm centres
- Absorption per baffle (both faces at 500 Hz): approximately 0.95 sabins
Arrangement: 8 rows of 6 baffles spanning the 10 m length of the room, with 1.25 m spacing between rows. This leaves clear sightlines to the instructor position and avoids the area directly above the main exercise zone.
Post-treatment RT60 (occupied):
A_total = 28.5 + (47 x 0.95) = 28.5 + 44.65 = 73.15 sabins
T60 = 0.161 x 360 / 73.15 = 57.96 / 73.15 = 0.79 seconds
This meets the target of 0.8 seconds and brings the studio into the range where instructor cues are intelligible even over loud music.
STI Improvement
At RT60 = 2.0 seconds with background music at 90 dBA and the PA system producing instructor speech at 85 dBA at the back of the room (8 m from the nearest ceiling speaker), the STI is approximately 0.30 — "poor."
At RT60 = 0.8 seconds with the same music level, the reverberant tail decays 2.5 times faster. The modulation depth of the speech signal is preserved more effectively, and the STI improves to approximately 0.52 — crossing the "fair" threshold. The improvement is not dramatic on the STI scale, but the perceptual difference is significant: participants at the back of the room transition from "cannot reliably understand words" to "can follow verbal cues with moderate effort."
For further improvement, reducing the music level by 3 dB (from 90 to 87 dBA) would push STI to approximately 0.58 — achieving "fair" to "good" intelligibility throughout the room.
Treatment Cost Analysis
| Item | Quantity | Unit Cost (GBP) | Total (GBP) |
|---|---|---|---|
| Mineral wool acoustic baffles (1200 x 600 x 50 mm) | 47 | £55–£85 | £2,585–£3,995 |
| Stainless steel wire suspension kits | 47 | £12–£18 | £564–£846 |
| Installation labour (1 day, 2-person team) | 1 | £800–£1,200 | £800–£1,200 |
| Subtotal — baffles | £3,949–£6,041 | ||
| Impact-resistant wall panels (optional, 25 mm PET felt, above 1.5 m) | 20 m² | £80–£120/m² | £1,600–£2,400 |
| Total with wall treatment | £5,549–£8,441 |
For an 80 m² studio, the cost per square metre is:
- Baffles only: £49–£76/m²
- Baffles + wall panels: £69–£106/m²
The Before-and-After: What Changes
Reverberation Time Comparison
| Frequency Band | Untreated (occupied) | Treated (occupied) | Reduction |
|---|---|---|---|
| 125 Hz | 2.8 s | 1.9 s | 32% |
| 250 Hz | 2.3 s | 1.2 s | 48% |
| 500 Hz | 2.0 s | 0.8 s | 60% |
| 1 kHz | 1.8 s | 0.7 s | 61% |
| 2 kHz | 1.5 s | 0.6 s | 60% |
| 4 kHz | 1.2 s | 0.5 s | 58% |
The treatment is most effective at mid and high frequencies (500 Hz–4 kHz), which are the frequencies most critical for speech intelligibility. At 125 Hz, mineral wool baffles are less effective — the wavelength (2.7 m) is much larger than the 50 mm baffle thickness, so the absorption coefficient drops to approximately 0.15–0.25 per baffle face. This is typical of thin absorbers and is generally acceptable in fitness studios because the primary goal is speech clarity rather than broadband reverberation control.
Music Clarity
The improvement is not limited to speech. Music clarity — the definition of individual instruments, bass lines, and rhythmic elements — also improves dramatically at RT60 below 1.0 seconds. In a reverberant room, the kick drum from one beat is still decaying when the next beat arrives, creating a muddy, indistinct sound. Reducing RT60 allows each musical element to be heard distinctly, which improves the energy and motivation of the class.
The metric for music clarity is C80 (clarity, defined as the ratio of early energy arriving within 80 ms to late energy arriving after 80 ms, per ISO 3382-1:2009 §4.5). In the untreated studio, C80 is approximately -2 dB (late energy dominates). After treatment, C80 improves to approximately +5 dB (early energy dominates). Values above +2 dB are generally considered "clear" for rhythmic music.
Common Mistakes in Gym Acoustic Treatment
Mistake 1: Acoustic Foam Tiles
Thin acoustic foam tiles (25–50 mm open-cell foam, typically with a wedge or pyramid profile) are the most commonly specified treatment for fitness studios — and the least effective. Standard acoustic foam has NRC of 0.40–0.55 when mounted directly on a wall, but the absorption is concentrated above 1 kHz. At 500 Hz, typical 50 mm foam provides α of only 0.20–0.30. At 250 Hz, it drops to 0.10–0.15.
The result is a room that sounds "dead" at high frequencies (the sibilance and cymbal crashes are absorbed) but still reverberant at mid frequencies (the voice and music fundamentals are not controlled). The echo is still there — it just sounds different. And the foam is vulnerable to impact damage, moisture, and cleaning chemicals, making it unsuitable for the gym environment.
Mistake 2: Treating Only the Ceiling
While the ceiling is the most effective single surface to treat (it has the largest unobstructed area and baffles provide dual-sided absorption), treating only the ceiling leaves the walls as efficient reflectors. In a narrow studio (8 m wide), sound reflects between the long walls with a round-trip time of approximately 47 ms (8 m x 2 / 343 m/s). This creates a flutter echo — a rapid series of distinct reflections — that is audibly distinct from diffuse reverberation and particularly distracting for speech.
Adding wall treatment on at least one of the long walls eliminates the flutter echo path. Impact-resistant PET (polyester) felt panels are ideal for this application: they withstand ball impacts and body contact, can be cleaned with standard gym cleaning products, and provide α of 0.60–0.80 at 500 Hz in 25 mm thickness.
Mistake 3: Over-Treating the Room
An RT60 below 0.4 seconds in a fitness studio creates the opposite problem — the room sounds uncomfortably dead. Music loses its energy and impact. The instructor's voice sounds flat and unnatural. Participants feel isolated rather than part of a group experience. The acoustic treatment should aim for 0.6–1.0 seconds, not the lowest possible value.
Sport-Specific RT60 Targets
Different fitness activities have different acoustic requirements:
| Activity | Primary Sound Source | Target RT60 | Reason |
|---|---|---|---|
| Spin / cycling | PA music + instructor mic | 0.6–0.8 s | High music levels, verbal cues critical |
| HIIT / boot camp | PA music + instructor voice | 0.6–0.8 s | Impact noise, fast verbal cues |
| Yoga / Pilates | Instructor voice + ambient music | 0.8–1.0 s | Quiet voice, intimate atmosphere |
| Boxing / martial arts | Instructor shouts + impact noise | 0.7–0.9 s | Very high impulse noise from impacts |
| Weight room (open floor) | Equipment + music system | 0.8–1.2 s | Speech less critical, music clarity wanted |
| Swimming pool | PA system + splash noise | 1.0–1.5 s | Hard surfaces, high humidity limits materials |
| Sports hall (multi-use) | PA + whistle + crowd | 1.2–1.8 s | Large volume, multiple simultaneous activities |
These targets are consistent with Sport England's "Acoustics in Sports Halls" guidance and with the general principles of ISO 3382-2:2008 for ordinary rooms.
The Business Case for Gym Operators
The 22% difference in member retention between acoustically treated and untreated studios translates directly to revenue. For a gym with 1,500 members paying an average of £45/month, a 22% improvement in retention for the subset of members who attend group classes (typically 40–50% of total membership) represents:
- 1,500 members x 45% class attendees = 675 class-attending members
- 22% of 675 = 149 members retained who would otherwise have left
- 149 members x £45/month x 12 months = £80,460 annual revenue protected
Related Reading
- The 125 Hz Problem Nobody Treats — why thin absorbers fail at low frequencies in any room type
- How Acoustic Panels Work: The Physics — the science behind absorption, reflection, and diffusion
- Acoustic Treatment Cost Calculator Guide — budgeting for acoustic treatment across building types