Amazon Music, Tidal and Deezer loudness: one master for all three
What is commonly reported for Amazon Music, Tidal and Deezer loudness, where the figures differ, and why one master at about −14 LUFS with true peak at or below −1 dBTP works across them.
Topic: Loudness for streaming and short video
Most artists who release on streaming want the same answer for every service: what level should the master be? Amazon Music, Tidal and Deezer each have a reputation for loudness handling, and the figures that circulate for them are not always the same. This article sets out what is commonly reported for each platform, where those reports differ, and why one master works across all three. The practical answer is short. Make one master near −14 LUFS with a true peak at or below −1 dBTP, and do not make a separate file for each platform.
The figures below are commonly reported references, not targets published by the platforms in the way Spotify publishes its own. Platforms can change their settings, so check the current behaviour before you rely on a number.
What is commonly reported
| Service | Commonly reported reference | Note |
|---|---|---|
| Amazon Music | about −14 LUFS | Commonly reported; not a figure we measured |
| Tidal | about −14 LUFS | Commonly reported for album playback |
| Deezer | about −15 LUFS | Commonly reported reference |
Two of the three figures are close to the −14 LUFS reference that Spotify publishes and many services use. Deezer's commonly reported −15 LUFS is one LU quieter, which matters only a little in practice. The main point is that all three sit within about one LU of the same region. That is why one master can cover them without a large gap.
Tidal's figure is most often reported for album playback, where tracks are normalized together. A single track in a playlist may sound different from the same track on an album. That is a playback detail, and it does not change the advice: a master near the reference is the safe choice for the whole catalogue.
Why one master is enough
A master that sits near −14 LUFS is turned down very little, or not at all, on services that normalize to about that reference. On a service that uses −15 LUFS as its reference, the difference is one LU, which is small enough that most listeners will not hear a change in level between two releases. On a service that does not normalize, the master plays at its own level, and a level close to the common reference keeps the track in line with other releases. The free streaming penalty calculator shows how much a given master is turned down on each platform.
What matters more than the exact figure is the true peak. Each platform and each encoder can add overshoot between samples. A true peak at or below −1 dBTP leaves room for that, and it is the one rule that every reference in this article points to. A separate master per service would not improve the result, and it would create more files to manage and more chances to upload the wrong one.
There is one exception. If a track is a deliberate outlier, such as a very quiet ambient piece or a loud club release, the best master may sit further from −14 LUFS. In that case, the right step is to choose the level for the music, then check the true peak and the playback, not to chase a number per platform.
The practical rule
- Master once, near −14 LUFS integrated. Aim for the middle of the common range, not the edges.
- Keep true peak at or below −1 dBTP. Check it on the final, exported file.
- Measure the exact file you will deliver. Use the free LUFS checker to read integrated loudness and true peak together.
- Deliver the same lossless master to every distributor. WAV or FLAC is the usual format for streaming delivery, and the services make their own compressed copies from it.
- Listen after release. Play the release on each service, or ask a few listeners to, and check for harshness on the loud parts.
For the wider picture of how streaming platforms handle loudness, read the streaming loudness guide. For Apple Music, which has its own handling and a commonly reported figure near −16 LUFS, read Apple Music loudness. For Spotify, read Spotify loudness.
Distributors usually take one audio file per release, and the same master can go to each of them. If you release through more than one distributor, send the same file to each, and keep the measurement with your release notes, so later checks compare like with like. When a service changes its behaviour, the measurement of your master does not change; only the playback does. That is one more reason to keep the master near the common reference rather than tuning it to one service's reported figure.
Common mistakes across these services
- Making a different master for each platform. One master at the common reference covers the three. Separate files add risk and do not buy a meaningful gain.
- Chasing the loudest master for one service. A file pushed far above −14 LUFS is turned down on services that normalize, and it loses dynamics for no net gain.
- Going quiet to be safe. A master much quieter than the reference sounds small next to other songs. Aim for a level that sounds finished.
- Trusting a number from a forum. Many sites repeat the same figures without a source. Treat the commonly reported figures here as guidance, and check the current platform behaviour before you release.
Where to start
Measure your final file with the LUFS checker. If it reads close to −14 LUFS and the true peak is at or below −1 dBTP, one master covers Amazon Music, Tidal, Deezer and the other common services. If it does not, a free AI mastering pass can bring it to a streaming level, and you can measure the result before you deliver it.
For a shorter look at the turn-down on each service, see the streaming loudness guide. For a similar comparison of smaller services, read Audiomack, Boomplay and JioSaavn loudness.
FAQ
What LUFS should a master be for Amazon Music, Tidal and Deezer?
Aim for about −14 LUFS integrated with a true peak at or below −1 dBTP. Commonly reported references are about −14 LUFS for Amazon Music and Tidal, and about −15 LUFS for Deezer. One master near −14 LUFS sits within about one LU of all three.
Should I make separate masters for each streaming service?
No, for most releases. One master near −14 LUFS with a true peak at or below −1 dBTP works across these services. Separate files add more chances for error and do not give a meaningful gain in level.
Why do the reported loudness figures differ?
Platforms set their own playback behaviour, and public reporting is not always consistent. Some figures are commonly reported for album playback rather than single tracks, and platforms can change their settings. Use the common reference as a guide and measure the file you deliver.
Check one master against all three services
The LUFS checker is free and needs no account. Measure your final file once, then release the same master everywhere.
Master a track free