LUFS in one paragraph
LUFS stands for Loudness Units relative to Full Scale. It's a measurement of how loud audio sounds to people, not how high its peaks go. Two tracks can both peak at 0 dBFS while one sounds twice as loud; their LUFS readings will show the difference. One LU is the same size as one dB, so a track at −10 LUFS is 4 dB louder than one at −14 LUFS.
How it's measured
The method comes from ITU-R BS.1770, with gating rules from EBU R128:
- K-weighting. The audio passes through a filter that cuts deep bass and lifts the upper mids a little, roughly matching the ear's sensitivity.
- Blocks. Loudness is measured in 400 ms blocks that overlap, starting every 100 ms.
- Gating. Blocks quieter than −70 LUFS are ignored, then blocks more than 10 LU below the average of what's left. That stops silence and quiet intros from dragging the figure down.
- Integrated loudness is the average of the blocks that pass. It's the one number platforms normalise to.
You'll also see short-term loudness (3-second windows), momentary loudness (400 ms), loudness range or LRA (how much the short-term level moves, in LU) and true peak in dBTP, which estimates the peaks between samples that appear once the audio is converted back to analogue or encoded to MP3 or AAC.
Why platforms turn your track down
Streaming services adjust playback volume so that a quiet folk song and a slammed EDM master sit at the same level. Spotify, for example, adjusts tracks to −14 LUFS by default. A master at −8 LUFS gets turned down by 6 dB, and a master at −20 LUFS gets turned up only as far as its peaks allow. The result: pushing a master louder than the target doesn't make it louder for listeners; it just costs dynamics and adds the risk of distortion.
Which target should I use?
| Destination | Integrated | True peak | Source |
|---|---|---|---|
| Spotify | −14 LUFS (Normal). Premium listeners can pick Loud −11 or Quiet −19. | Below −1 dBTP; below −2 dBTP if louder than −14 LUFS | Spotify for Artists |
| Podcasts (Apple) | About −16 LUFS ±1 | −1 dBTP max | Apple Podcasts for Creators |
| European broadcast | −23 LUFS | −1 dBTP max | EBU R128 |
| US broadcast | −24 LKFS | −2 dBTP | ATSC A/85 |
YouTube, Apple Music, Tidal and Amazon also normalise, but don't all publish an exact target; engineers' measurements generally put them between about −14 and −16 LUFS. If you deliver one master everywhere, around −14 LUFS with true peak at −1 dBTP is a safe middle ground for music.
Practical advice
- Music: master for how it sounds, then check. Between −14 and −9 LUFS is normal for modern pop and dance; just keep true peak at −1 dBTP or lower so lossy encoding doesn't clip.
- Podcasts and voice: aim for −16 LUFS. Speech at −14 or louder is tiring to listen to.
- Broadcast and film: follow the delivery spec exactly; −23 or −24 is a requirement, not a suggestion.
- Comparing mixes: always compare at matched loudness. Louder almost always sounds "better" for a few seconds.
Measure your own file
WAVR's loudness meter reports integrated, short-term and momentary loudness, LRA and true peak, and tells you how many dB to add or remove to hit each target. It runs in your browser, so unreleased masters aren't uploaded. In the editor, the A/B preview with Match loudness lets you compare versions at the same LUFS.