Glossary
LUFS: the loudness number YouTube and Spotify actually use
Your video can look perfect and still feel amateur if the volume jumps around or arrives whisper-quiet. LUFS is the measurement platforms use to judge loudness — and they will turn your audio up or down to match. This guide explains LUFS in plain language and shows how to master to it.
From the Mellius product team · Updated September 28, 2026
Conspiracy theorists, you may hate this
AI Director
›
Just type what you need
- ›Remove all silent parts
- ›Add bold kinetic captions
- ›Only have Bill Ackman speaking
- ›Cut to highlights only
- ›Add cinematic transitions
- ›Match cuts to the beat
The short answer
LUFS (Loudness Units Full Scale) is the standard way to measure how loud audio feels to ears over time, not peak levels. Streaming platforms normalize everything to a target — around minus 14 LUFS for YouTube and Spotify, minus 16 for podcasts — turning louder uploads down. Mastering near the target keeps your volume consistent without pumping or distortion.
It measures perceived loudness
LUFS averages loudness the way ears hear it, weighting frequencies and time — unlike peak meters, which only catch the loudest instant.
Platforms normalize to targets
YouTube, Spotify, and podcast apps adjust playback toward their house target, so mastering near it means your mix plays back as intended.
Consistency beats maximum volume
A steady integrated level with controlled peaks sounds louder and cleaner than a crushed mix the platform turns down anyway.
What LUFS actually measures
Peak level tells you the single loudest moment; LUFS tells you how loud the whole thing feels. The integrated LUFS value averages loudness across your entire video with perceptual weighting, so bass-heavy rumble counts less and midrange speech counts more — roughly matching human hearing. Short-term and momentary LUFS readings show loudness over seconds and fractions of a second, which is how you spot the jumps between quiet dialogue and blaring intros.
Two mixes can share identical peak levels yet feel wildly different in volume — that gap is exactly what LUFS quantifies. A dynamic podcast with quiet passages might peak at the ceiling yet integrate at minus 20, while a compressed ad peaks identically but integrates at minus 10. Platforms normalize by integrated LUFS, so the ad gets turned down and the podcast gets turned up.
The targets to master to
For YouTube and music streaming, aim for integrated loudness around minus 14 LUFS with true peaks below minus 1 dBTP. Podcasts conventionally target minus 16 LUFS stereo (minus 19 mono). Broadcast television in many regions standardizes on minus 23 or minus 24. These are not laws — platforms accept a range — but landing near the target means normalization barely touches your mix.
Chasing louder-than-target masters backfires: the platform simply turns them down, and your over-compressed dynamics stay squashed while quieter masters sound punchier by comparison. The winning strategy is moderate compression, clear dialogue, and peaks with headroom. Leave the final loudness to the platform's normalizer instead of fighting it.
Beginner loudness mistakes
Mixing by waveform height is the classic error: waveforms show peaks, not perceived loudness, so a tall-but-thin music bed can measure quieter than short-but-dense dialogue. Use a LUFS meter on your master bus and mix dialogue to sit consistently, then fit music and effects around it. If viewers ride the volume control, your integrated balance — not your peaks — is wrong.
The second mistake is slamming a limiter to win the loudness war on a platform that normalizes. Crushed transients, pumping music, and distorted sibilance all survive normalization while the extra volume does not. A gentler master with intact dynamics normalizes to the same playback level and sounds dramatically better getting there.
A practical sequence
Put it into practice
- 01
Put a LUFS meter on your master
Add a loudness meter to the master bus and play the whole video through. Note the integrated LUFS and the true peak values against your platform's target.
- 02
Balance dialogue first, then everything else
Set clean, consistent dialogue levels with light compression, then tuck music and effects underneath. Re-measure; adjust clip gain rather than crushing the master limiter.
- 03
Export with headroom and re-check
Keep true peaks below minus 1 dBTP and integrated loudness near target, then meter the export itself. In Mellius, you can ask for the music to sit under the voice in plain language, then fine-tune the balance on the editable timeline with undo.
Frequently asked questions
What LUFS should I master YouTube videos to?
Aim for around minus 14 LUFS integrated with true peaks below minus 1 dBTP, which is YouTube's normalization target. Content slightly quieter or louder still plays fine — YouTube adjusts it. Prioritize consistent dialogue and clean peaks over hitting the number exactly.
What is the difference between LUFS and decibels?
Decibels measure signal level at an instant, including momentary peaks. LUFS uses decibel units but applies perceptual weighting and time averaging to estimate how loud something feels over time. Think of dB as the speedometer and LUFS as the average speed of the whole trip.
Do I need a special plugin to measure LUFS?
Most modern editors and audio tools include loudness meters, and capable free meters exist for every platform. You need integrated, short-term, and true-peak readouts at minimum. Measure the full program from start to finish rather than judging by a loud section.
Continue reading
Primary sources
Product capabilities change. These official references support the comparisons above.
Describe the edit. Keep the timeline.
Mellius turns conversational direction into cuts, captions, graphics, motion, color, and sound you can still inspect and change.
Start editing free