How to Read an Acoustic Report — What Every Architect Needs to Understand

1 in 4 acoustic reports submitted for building certification contains at least one methodological error that could change the compliance verdict — according to a 2023 analysis of 340 WELL and BREEAM acoustic assessments reviewed by UK acoustic consultancies. The most common errors are insufficient measurement positions (violating ISO 3382-2:2008 §5), measurements taken under non-representative HVAC conditions, and application of the wrong compliance standard. Architects and project managers who cannot read an acoustic report critically are unable to spot these errors, and end up either accepting a false "fail" (triggering unnecessary remediation) or a false "pass" (storing up complaints for occupancy).

This guide walks through every section of a typical acoustic report, explains what the numbers mean, shows what "pass" and "fail" look like with real examples, and identifies the red flags that should prompt you to question the results.

Report Structure: The Seven Sections

A professional acoustic report following ISO 3382 methodology typically contains these sections:

1. Executive summary — pass/fail verdicts for each room and metric
2. Scope and standards — what was measured, which standards apply
3. Measurement methodology — equipment, positions, procedure
4. Results: Reverberation time (RT60) — tables and graphs by room
5. Results: Background noise — dBA levels and NC/NR curves
6. Results: Speech intelligibility (STI) — where required
7. Compliance matrix — room-by-room pass/fail against each criterion

Section 1: Executive Summary

What to Look For

The executive summary should state:

• The total number of spaces tested
• The number that pass and fail each criterion
• The specific criterion used (not just "the room passes" — passes what?)
• Any caveats or limitations

Red Flag: Vague Compliance Statements

An executive summary that says "the rooms generally meet acoustic requirements" without specifying which requirements, which standard, and which rooms is professionally inadequate. Every room should have a clear pass or fail against a specific, named criterion with a numeric threshold.

Good: "Meeting Room 3A achieves RT60 = 0.52 s (500-2000 Hz average). This meets WELL v2 Feature 74 Part 1 (threshold: ≤ 0.6 s). Pass."

Bad: "Meeting Room 3A has acceptable acoustic performance."

Section 2: Scope and Standards

This section lists which standards the acoustic consultant has assessed against. Verify that these are the correct standards for your project. Common mismatches:

If the report references the wrong standard, every compliance verdict is suspect. This is not a minor administrative error — different standards have different thresholds, different frequency ranges, and different measurement protocols.

Section 3: Measurement Methodology

Equipment

The report should list:

• Sound source: Omnidirectional loudspeaker (for interrupted noise method) or starter pistol/balloon (for impulse response method). Per ISO 3382-2:2008 §5.2, the source must be omnidirectional within ±1 dB in any direction.
• Microphone: Omnidirectional measurement microphone with known frequency response. Class 1 per IEC 61672-1.
• Sound level meter / analyser: With octave-band analysis capability. Must be calibrated per IEC 61672-3.
• Calibrator: Pistonphone or acoustic calibrator, calibration certificate current within 12 months.

Red Flag: No Calibration Dates

If the report does not include calibration certificate references or dates, the measurement data is unreliable. A microphone that has not been calibrated for three years may have sensitivity drift of ±2 dB, which translates to RT60 measurement error of approximately ±15%.

Measurement Positions

Per ISO 3382-2:2008 §5.3, measurement positions must satisfy:

• Source-receiver distance ≥ 1.0 m
• Source-receiver distance ≥ 2 × critical distance (d_c)
• Microphone at least 1.0 m from any reflective surface
• Microphone height: 1.2 m ± 0.1 m (seated ear height) for occupied-mode rooms, or 1.5 m for standing-mode rooms
• Minimum 3 source-receiver combinations for rooms < 300 m²

Red Flag: All Microphones in One Area

If all three microphone positions are clustered near the front of a meeting room (close to the acoustic ceiling absorption), the measured RT60 will be artificially low. Positions should be distributed throughout the room to capture the spatial average.

Section 4: Reverberation Time (RT60)

Understanding the RT60 Table

A typical RT60 results table looks like this:

What the Numbers Mean

• 125 Hz: Bass reverberation. Values above 0.8 s indicate a "boomy" room. Most standards do not set limits at 125 Hz, but subjective quality suffers when the 125 Hz RT60 is more than 1.5× the 500 Hz value.
• 250 Hz: Low-mid reverberation. This frequency range is responsible for "muddy" or "warm" room sound. Important for speech clarity.
• 500-2000 Hz: The speech frequency range. Most standards set their RT60 targets as an average across these three octave bands. This is the primary compliance criterion.
• 4000 Hz: High-frequency reverberation. In small rooms this is usually well-controlled because air absorption and carpet provide significant attenuation. Values above 0.6 s at 4000 Hz in rooms under 200 m³ are unusual and may indicate a measurement error.
• Avg 500-2k: The arithmetic mean of the 500, 1000, and 2000 Hz values. This is the single number most commonly compared against the standard's threshold.

Red Flag: T20 vs T30 vs T60

RT60 is defined as the time for sound to decay by 60 dB. In practice, achieving a 60 dB dynamic range in a measurement is difficult in occupied buildings (ambient noise limits the measurable decay range). Acoustic consultants therefore measure T20 (decay from -5 dB to -25 dB, extrapolated to 60 dB) or T30 (decay from -5 dB to -35 dB, extrapolated to 60 dB) per ISO 3382-2:2008 §4.

T20 and T30 should agree within 10%. If they differ by more than 10%, the sound field is not decaying exponentially, which indicates either non-diffuse conditions (common in rooms with concentrated absorption), a high ambient noise floor cutting into the decay curve, or insufficient source level.

The report should state which parameter was used (T20 or T30) and whether the linearity check was satisfied (r² ≥ 0.95 for the least-squares fit to the decay curve). If this information is absent, the RT60 values may be unreliable.

Worked Example: Checking the Boardroom

The boardroom in the table above shows RT60 = 0.71 s (500–2000 Hz average) against a target of ≤ 0.7 s. This is a marginal fail — 0.01 s above the threshold. Before accepting this verdict:

1. Check the measurement uncertainty: ISO 3382-2 notes that RT60 measurement uncertainty is approximately ±5% for well-controlled measurements with adequate source-receiver positions. A measured value of 0.71 s has a 95% confidence interval of approximately 0.67–0.75 s. The true RT60 could be below 0.7 s.

1. Check the measurement conditions: Was the room empty or furnished? If measured empty, adding 20 people (the typical boardroom occupancy) would add approximately 10 m² Sabine of absorption, reducing the RT60 by approximately 0.05–0.08 s. The occupied RT60 would be approximately 0.63–0.66 s — a pass.

1. Check the individual frequency values: The 2000 Hz value (0.72 s) is higher than the 500 Hz value (0.74 s) but lower than 1000 Hz (0.68 s). This slight increase at 2000 Hz is unusual and may indicate a measurement artefact (insufficient decay range at that frequency due to ambient noise) or a genuine acoustic anomaly (reflective surface at ear height).

Section 5: Background Noise

Understanding Background Noise Results

Background noise is typically reported as:

• Leq dBA: The equivalent continuous A-weighted sound pressure level over the measurement period (usually 5–15 minutes)
• Octave-band spectrum: Sound pressure levels at each octave band from 63 Hz to 8000 Hz
• NC/NR rating: The Noise Criteria (US) or Noise Rating (Europe) curve that the measured spectrum matches or falls below

• Flat spectrum: 30 dB at every octave band from 125 to 4000 Hz. This sounds like "white noise" — uniform, not objectionable.
• Low-frequency heavy: 45 dB at 63 Hz, 40 dB at 125 Hz, 30 dB at 250 Hz, 25 dB at higher frequencies. This sounds like a low rumble (typical of HVAC fan noise or transformer hum). The dBA value is 35, but the subjective quality is poor.

Red Flag: Measurement During Non-Typical Conditions

Background noise measurements must represent typical operating conditions. Common pitfalls:

• Weekend measurement with HVAC on night setback mode (artificially quiet)
• Measurement during fire alarm testing or building works (artificially loud)
• Open windows during measurement (external noise dominates)
• HVAC system not yet commissioned at time of measurement

Section 6: Speech Intelligibility (STI)

When STI Is Required

STI measurements are required by:

• WELL v2 Feature 74 Part 3 (speech privacy in open-plan offices)
• BB93:2015 (UK schools — STI ≥ 0.60 in classrooms)
• ANSI S12.60-2010 (US schools — references STI indirectly via RT60 and background noise limits)

Understanding STI Results

STI ranges from 0.00 (completely unintelligible) to 1.00 (perfect intelligibility). The IEC 60268-16:2020 quality scale:

For speech privacy applications (WELL Part 3), the goal is low STI — you want speech to be unintelligible between workstations. For classrooms and lecture halls, the goal is high STI — you want every word clearly understood.

Red Flag: RASTI Instead of STIPA

RASTI (Rapid STI) was defined in an older version of IEC 60268-16 and uses only two octave bands (500 Hz and 2000 Hz). It has been withdrawn from the current standard and is no longer accepted by WELL or BB93. If the report uses RASTI, the STI data is obsolete and must be re-measured using STIPA (which uses all seven octave bands from 125 Hz to 8000 Hz).

Section 7: Compliance Matrix

The compliance matrix is the summary table that maps every room against every applicable criterion. It should look like this:

How to Read the Matrix

Scan the "Fail" entries first. For each failure:

1. How far from the threshold? A room at 0.62 s against a 0.60 s target is a marginal fail that may be resolvable with furnishing changes. A room at 0.95 s against a 0.60 s target requires significant intervention.

1. What is causing the failure? Cross-reference with the detailed results. If RT60 fails at 125 Hz but passes at 500–2000 Hz, the problem is bass reverberation. If background noise fails at 250 Hz but passes at higher frequencies, the problem is likely HVAC fan noise or duct rumble.

1. Is the failure genuine? Check the measurement methodology for errors. A marginal fail (within the measurement uncertainty band) warrants a request for additional measurements or a challenge based on occupied vs unoccupied conditions.

Five Questions to Ask About Any Acoustic Report

1. What standard was used? Verify it matches your project's certification and regulatory requirements.
2. How many measurement positions? Check against ISO 3382-2 minimums (3 source-receiver combinations for rooms < 300 m²).
3. What were the HVAC conditions? Confirm measurements represent normal occupied operation.
4. Are calibration certificates current? Equipment must be calibrated within 12 months of measurement.
5. Was T20 or T30 used, and did the linearity check pass? If not stated, the RT60 values may be unreliable.

Related Reading

• The 8-Step Acoustic Design Process — the full workflow from brief to handover, including the measurement step
• How to Calculate RT60 Step by Step — understand the numbers behind the report's RT60 predictions
• RT60 Measurement Methods Comparison — interrupted noise vs impulse response vs MLS