What Is LUFS?
LUFS stands for Loudness Units relative to Full Scale. It is a standardized measurement of perceived loudness defined by the ITU-R BS.1770 specification, which was adopted by the European Broadcasting Union (EBU) and is now used by virtually every streaming platform in the world.
Unlike peak meters that show the highest instantaneous signal level, LUFS measures how loud audio actually sounds to human ears over time. It accounts for the fact that our hearing is more sensitive to midrange frequencies than to very low or very high ones. A track peaking at 0 dBFS might measure -8 LUFS or -16 LUFS depending on its density, arrangement, and frequency content.
LUFS is the reason modern streaming services can play songs from different decades, genres, and production styles at roughly the same perceived volume without any track blowing out the listener's ears or disappearing into silence.
LUFS vs dB: Understanding the Difference
Decibels (dB) and LUFS both describe audio levels, but they measure fundamentally different things:
- dBFS (decibels relative to Full Scale) measures the instantaneous electrical level of a digital audio signal. It tells you how close the signal is to clipping (0 dBFS). It does not account for how loud the audio actually sounds.
- LUFS measures perceived loudness over time, weighted by a frequency curve (K-weighting) that approximates human hearing. Two signals can have the same peak dBFS but wildly different LUFS readings if one is sparse and the other is dense.
Example: A solo acoustic guitar track might peak at -3 dBFS but measure -20 LUFS integrated. A heavily compressed EDM track might also peak at -3 dBFS but measure -7 LUFS integrated. Same peaks, completely different perceived loudness. LUFS captures that difference. Peak dBFS does not.
There is also dBTP (decibels True Peak), which accounts for inter-sample peaks that can occur during digital-to-analog conversion. Most streaming platforms recommend a true peak ceiling of -1 dBTP (some suggest -2 dBTP) to prevent distortion during encoding.
Integrated, Short-Term, and Momentary Loudness
LUFS is not a single number. There are three measurement windows, and each serves a different purpose in mastering:
- Integrated LUFS is the average loudness across the entire duration of the track. This is the number streaming platforms use for normalization. When someone says a track is "-14 LUFS," they mean the integrated measurement.
- Short-term LUFS measures loudness over a rolling 3-second window. It is useful for identifying sections of a song that are significantly louder or quieter than the average, helping you spot consistency problems.
- Momentary LUFS measures over a 400-millisecond window. It captures transient loudness spikes and is helpful for monitoring real-time dynamics during playback.
When mastering for streaming, your primary target is the integrated LUFS reading. But keeping an eye on short-term and momentary values helps ensure that no section of your track feels dramatically out of balance with the rest.
Streaming Platform Loudness Targets
Every major streaming service normalizes audio to a specific loudness target. Here are the current standards:
- Spotify: -14 LUFS integrated, -1 dBTP true peak. Spotify turns down tracks louder than -14 LUFS and can optionally turn up quieter tracks depending on the user's normalization setting.
- Apple Music: -16 LUFS integrated, -1 dBTP. Apple's Sound Check normalization is slightly more conservative, favoring dynamic masters.
- YouTube: -14 LUFS integrated, -1 dBTP. YouTube normalizes loudness for all audio content, including music videos and podcasts.
- Tidal: -14 LUFS integrated, -1 dBTP. Tidal follows the same standard as Spotify for its normalization algorithm.
- Amazon Music: -14 LUFS integrated, -2 dBTP. Amazon applies a slightly stricter true peak limit.
The practical takeaway: if you master your music to -14 LUFS integrated with a true peak ceiling of -1 dBTP, you will be well within the safe range for every major platform. Masters louder than -14 LUFS will be turned down, which wastes the dynamic range you sacrificed to get there. Masters quieter than -14 LUFS may be turned up (depending on the platform), which is generally fine as long as the track sounds intentionally dynamic.
The Loudness War: Did LUFS End It?
From the late 1990s through the early 2010s, the music industry was locked in a loudness war. Engineers pushed masters louder and louder, crushing dynamic range with heavy limiting, because louder tracks sounded more impressive in direct comparison. Albums from this era routinely measured -6 to -4 LUFS, with some hitting -3 LUFS or higher.
The cost was severe. Over-compressed masters sounded fatiguing, distorted, and lifeless. Quiet passages disappeared. Transients were flattened. Musical dynamics, the very thing that gives a song its emotional impact, were destroyed in the pursuit of raw volume.
Loudness normalization changed everything. When Spotify, Apple Music, and YouTube began automatically turning down loud tracks to their target level, the reason to over-compress mostly disappeared. A track mastered at -6 LUFS gets turned down to -14 LUFS on Spotify, and now it sounds worse than a track mastered at -14 LUFS with its full dynamic range intact.
In theory, normalization ended the loudness war. In practice, our own measurements say it hasn't. We measured 151 commercially released masters from 2004 to 2024 (most from 2018 on) across 17 genres, using the same ITU-R BS.1770-4 measurement the streaming services use, plus 4× oversampled true-peak detection:
| Median integrated loudness | -10.6 LUFS |
| Louder than Spotify's -14 LUFS target | 81% |
| Louder than Apple Music's -16 LUFS target | 92% |
| Median amount Spotify turns them down | 4.2 dB |
| True peaks above 0 dBTP (clipping before encoding) | 62% |
Commercial releases are still mastered 3 to 4 dB louder than Spotify plays them, and most of that extra loudness is simply turned back down. The cost shows up in punch: the loudest masters (above -9 LUFS) had a median crest factor about 2 dB lower than masters between -12 and -9 LUFS.
Loudness also varies a lot by genre. In our 2026 loudness research, metal had the quietest median (about -14 LUFS) and bedroom pop and pop the loudest (about -8 LUFS). See how much Spotify and Apple Music turn down each genre. So the real lesson is not "always hit -14". Master to a level that suits the song, keep true peak at or below -1 dBTP, and don't trade away punch for loudness that the platforms will remove anyway. See what Spotify does to loud masters and what Apple Music's Sound Check does.
Check your track free: measure its LUFS and true peak and see what Spotify, Apple Music, YouTube and 5 more platforms will do to it. Open the free LUFS checker →
How to Measure LUFS
You need a loudness meter that conforms to the ITU-R BS.1770 standard. There are several ways to get one:
- DAW built-in meters. Most modern DAWs (Logic Pro, Ableton Live 12, Pro Tools, Studio One) include LUFS metering in their master bus or metering plugins. Check your DAW's documentation for where to find it.
- Free plugins. Youlean Loudness Meter is the most popular free option. It provides integrated, short-term, and momentary LUFS readings along with a loudness history graph. dpMeter and MLoudnessAnalyzer are other free choices.
- Online mastering platforms. LuvLang displays real-time LUFS metering during the mastering process, so you can see exactly where your track lands relative to streaming targets before you export.
When measuring, always let the full track play from start to finish before reading the integrated LUFS value. Pausing or skipping ahead will give you an inaccurate number. The integrated measurement gates silence automatically (per the BS.1770 spec), so gaps between sections will not drag the number down artificially.
Practical Tips for Hitting Your Target
Reaching the right LUFS target is not about slamming a limiter until the meter reads -14. It is about building loudness gradually through the entire mastering chain. For a full walkthrough of gain staging, compression, and limiting techniques, see our guide on how to make your song louder without destroying it. Here is how to do it correctly:
- Start with a good mix. A well-balanced mix with proper gain staging will reach -14 LUFS with minimal limiting. See our mastering prep checklist for how to prepare your mix. If you need more than 3 to 4 dB of gain reduction on your limiter, the mix likely needs work first.
- Use gentle compression before the limiter. A bus compressor with a low ratio (2:1 or less) and slow attack can add 1 to 2 dB of density without squashing transients. Multiband compression lets you tame specific frequency ranges independently.
- Set your true peak ceiling at -1 dBTP. This gives encoding headroom for lossy formats like MP3, AAC, and Ogg Vorbis that streaming platforms use. Without this headroom, inter-sample peaks can cause clipping during encoding.
- Compare at matched loudness. When A/B testing your master against a reference track, use a loudness-matched comparison. Our ears perceive louder audio as better, so an unmatched comparison will always favor whichever signal is louder, regardless of quality.
- Trust the meter, not your ears alone. After hours of listening, your perception adapts. The LUFS meter gives you an objective reference point. If the meter reads -14 LUFS and the track sounds full and dynamic, you are done.
Understanding LUFS is not just technical knowledge. It directly affects how your music sounds to every listener on every platform. Get it right, and your masters will translate everywhere. Get it wrong, and the streaming platform's algorithm will make the correction for you, usually in a way that does not flatter your mix.
Part of the audio mastering guide: read every guide in order.
Check your own bounce with the free LUFS checker, and see how the dynamic range compares with the dynamic range checker.