Guide
LUFS puts a number on loudness, weighted the way ears actually work
LUFS stands for Loudness Units relative to Full Scale. It is the unit modern loudness is measured in, weighted to match how human hearing works rather than what a meter needle sees. One figure, integrated LUFS, describes the average loudness of a whole file. This page defines the terms; the delivery targets for a feed have a page of their own.
- LUFS integrated, the common feed landing zone
- -14
- dBTP, the peak ceiling that prevents crackle
- -1
- Readings on a loudness meter worth knowing
- 3
Why decibels were not enough
A peak meter reports the highest point a waveform reaches. Two files can peak identically while one sounds twice as loud as the other, because loudness lives in the average energy, not the maximum.
LUFS measures that average, filtered to weight the frequencies ears care about. A rumble that a meter sees but a listener barely hears counts for less.
The practical result is a number that tracks perception. If file A reads three units above file B, listeners will agree A is noticeably louder.
The three readings you will meet on a meter
Momentary and short-term loudness describe the last fraction of a second and the last few seconds. They move constantly and are for watching while you work.
Integrated loudness is one number for the entire file, and it is the one platforms and delivery specs talk about. When someone says this video is -14, they mean integrated.
Loudness range, LRA, describes how far apart the loud and quiet stretches sit. A wide range sounds dynamic and forces volume-riding; a narrow one sounds steady.
The readings, from twitchiest to most quoted
Every reading comes from the same measurement standard. Only integrated describes the file as a whole.
Momentary
The last 400 ms
Short-term
The last 3 seconds
Integrated
The whole file, one number
LRA
Loud against quiet, as a spread
This is the whole editor
Highlight a phrase and a clip lands on those exact words. No timeline, no keyframes, no layers.
The numbers that matter for a vertical feed
Short-form video clusters around -14 LUFS integrated, with true peaks held under -1 dBTP so the encode does not clip. Those two numbers cover different failures: one is being quieter than the feed, the other is crackle.
The pair travels together. A file at -14 with clipped peaks sounds broken, and a clean file at -22 sounds weak next to whatever plays before it.
Every Cutroom export is normalized to -14 LUFS with peaks held under the ceiling, so the measurement happens whether or not anyone reads the meter.
Questions people ask
- Is LUFS the same as dB?
- They share the scale's shape but measure different things. dB in a peak meter is an instantaneous maximum. LUFS is perceived loudness averaged over time with hearing-weighted filtering. A file needs both looked at.
- What LUFS should a vertical video be?
- Aim for -14 integrated with true peaks under -1 dBTP. No platform publishes an official figure, and -14 is the middle of where measured feed content lands rather than a spec anyone enforces.
- Can I hear the difference between -14 and -13?
- Not reliably. One loudness unit corresponds to about one decibel of level change, which is around the edge of perception on programme material. The gap that matters is six or eight units, not one.
One number for the average, one for the ceiling, and a meter to read them on. Or export from a pipeline that sets both without asking, which is what ours does.