The Short Answer
Almost every genre is mastered louder than Spotify plays it. In our measurements, the median release in 16 of 17 genres was louder than Spotify's -14 LUFS target, so Spotify turns it down. Only metal came in below it. On Apple Music, which normalizes to -16 LUFS, every genre's median release gets turned down.
The loudest genre, bedroom pop, loses a median 5.6 dB on Spotify and 7.6 dB on Apple Music. That's loudness the artist paid for in punch and then never got to use.
Loudness and Normalization by Genre
Each row is the median of the releases we measured in that genre. "Turns down" is how far Spotify (-14 LUFS) and Apple Music (-16 LUFS) lower the median release. Sorted quietest to loudest.
| Genre | Releases | Median LUFS | Spotify turns down | Apple Music turns down | Above -1 dBTP |
|---|---|---|---|---|---|
| Metal | 10 | -14.7 | 0.0 dB | 1.3 dB | 40% |
| R&B | 8 | -13.0 | 1.0 dB | 3.0 dB | 50% |
| Indie | 8 | -12.7 | 1.3 dB | 3.3 dB | 38% |
| Rock | 8 | -12.4 | 1.6 dB | 3.6 dB | 38% |
| Classical | 8 | -12.2 | 1.8 dB | 3.8 dB | 100% |
| Country | 10 | -11.6 | 2.4 dB | 4.4 dB | 50% |
| Jazz | 9 | -11.2 | 2.8 dB | 4.8 dB | 56% |
| EDM | 10 | -11.0 | 3.0 dB | 5.0 dB | 50% |
| Hip-hop | 10 | -10.7 | 3.3 dB | 5.3 dB | 90% |
| Latin | 10 | -10.7 | 3.3 dB | 5.3 dB | 60% |
| Electronic | 10 | -10.2 | 3.8 dB | 5.8 dB | 70% |
| Acoustic | 8 | -10.2 | 3.8 dB | 5.8 dB | 100% |
| Pop | 10 | -10.1 | 3.9 dB | 5.9 dB | 60% |
| Gospel | 10 | -9.9 | 4.1 dB | 6.1 dB | 100% |
| Folk | 9 | -9.7 | 4.3 dB | 6.3 dB | 100% |
| Reggae | 5 | -9.3 | 4.7 dB | 6.7 dB | 100% |
| Bedroom pop | 8 | -8.4 | 5.6 dB | 7.6 dB | 88% |
Samples are 5 to 10 releases per genre, so read each row as indicative, not definitive. The overall figures (151 releases) are more reliable than any single genre.
Five Things the Data Shows
1. The loudness war isn't over, it just moved. The median release across all genres sits at -10.6 LUFS, about 3.5 dB louder than Spotify plays it. Genres built on intimacy, like bedroom pop, acoustic and folk, are now among the loudest.
2. Metal is the quiet one. Metal's median (-14.7 LUFS) sits just under Spotify's target, and it had the second-highest crest factor in our sample, just behind classical (about 14.7 dB between peaks and average level). Heavy music doesn't have to be crushed.
3. True peak is the bigger problem. In five genres (acoustic, classical, folk, gospel and reggae) every release we measured peaked above -1 dBTP, the ceiling Spotify and Apple recommend to avoid encoding distortion. Hip-hop came in at 90%. Those peaks stay in the file even after normalization turns the level down, so lossy encoding can still distort them.
4. Classical proves loudness and peaks are different problems. Classical releases were relatively quiet (median -12.2 LUFS) yet every one peaked above -1 dBTP, because dynamic music has big transient peaks. A true-peak limiter matters even when you aren't chasing loudness.
5. Loud masters lose punch they never get back. Across all 151 releases, masters louder than -9 LUFS had a median crest factor about 2 dB lower than masters between -12 and -9 LUFS. Normalization lowers the volume, but it doesn't restore those transients.
What This Means for Your Master
- Don't chase your genre's loudest release. Anything above -14 LUFS is turned down on Spotify and above -16 LUFS on Apple Music.
- Master to suit the song. Somewhere between -14 and -10 LUFS works for most genres; very dynamic music can sit lower.
- Always set a true-peak ceiling of -1.0 dBTP. The data shows this is where most commercial masters fall short.
- Remember the exceptions. Bandcamp doesn't normalize at all (what that means), and Tidal only turns tracks down (details).
How We Measured
We analyzed 151 commercially released masters from our research library (released 2004 to 2024, most from 2018 on) across 17 genres. Integrated loudness was measured with ITU-R BS.1770-4, the same standard streaming services use, and true peak with 4× oversampled detection. Turn-down figures apply each platform's published target to the median release in each genre; real playback also depends on listener settings and, on Tidal and Apple Music, album-level normalization. Full methodology and more findings are in our 2026 loudness research.
Using this data? You're welcome to cite it with a link to this page.