Weekly Tip #29: Avoid Normalizing
When working in Photoshop… Do you resize the picture every now and then, just because you can?
I doubt it! Yet people keep normalizing their audio every chance they get.
Makes no sense.
There are a couple of situations where normalizing comes in handy. But these are rare occasions in the music production process.
Let me first tell you what normalizing is in a nutshell and when NOT to normalize.
Peak normalizing is the process of finding the loudest peak of the audio signal, and applying a constant amount gain to the entire signal so that the highest peak reaches the target level.
In practice it’s often (and often unnecessarily) done to bring the signal to 0 dBFS – it’s maximum digital level.
Virtually every DAW and audio editing software out there has a built-in function for doing this automatically.
So, what’s the fuss?
People like to normalize for the wrong reasons.
Before you normalize, realize that:
- It doesn’t affect the dynamic range of the signal in any way. Once again: You won’t gain more dynamic range by normalizing. The difference between the loudest and quietest parts of the signal stays the same.
- Normalizing every sound/track in your mix won’t make it sound louder or better.
- You may risk creating inter-sample peaks (and thus, clipping when the sound goes to a poor DAC).
- It will bring up the noise by the same amount.
- You lose headroom and risk clipping when the audio with no headroom is processed with certain plugins.
- Any processing step you take degrades the original sound – the less you process, the better.
Nearly always in music production and mixing situations, normalizing is not really necessary.
You’re better off concentrating your efforts recording/bouncing things loud enough (-3 to -6dBSF) in the first place.
The exceptions? There are two kinds of situations where normalization is required:
- If the signal is so quiet that the volume controls in your DAW are not enough to bring it up to par with the other elements in the mix.
- If you must make the signal peak at a certain predefined level (for example when doing sound design, mastering or working with test tones).
If you can think of more, do let us know in the comments!
![]() |















How about normalizing the final bounced mix output in logic?
No reason to do that – just bounce final mix at appropriate level (usually this means leaving 3 to 6 dB of headroom).