Or, I’m an Artist: How Loud Should my Recording Be?

A friend of mine, Jason Sibi Okumu, once told me that if you sing quietly, people will lean in to listen. Lowland Hum used to post with the hashtag #SupportQuietMusic. Bon Iver’s first record measures at an uncharacteristically quiet level for the 2000s.

It’s not just a folkie thing. When it comes to making records, quite literally, nothing is loud if everything is loud. So how loud should your record be? (The jargony TL;DR short answer, at least the way I mix and master and recommend: between -14 to -8 LUFS, give or take.)

If you’re an artist, especially one new to recording, whether you’re working with an engineer or DIYing, I hope this article gives some insight into the process of mixing and mastering for loudness.

If you, like me, have ever recorded yourself in GarageBand, then exported it to listen in iTunes alongside your favorite tracks, then felt frustrated and sad that it was so quiet, only to boost the volume of your track in the DAW and find out now it’s distorting like hell, you’ve likely wondered: there must be some way to make this louder and do it right.

To this end, audio engineers use compression and its OP younger sibling, limiting. Put simply, compressors do to the waveform what it says on the tin: they compress the highest peaks of the waveform, which allows you to turn up the overall gain (loudness) of the track. Compression is often used serially throughout the mix, not uncommonly on every channel, as needed. Most of the time compression sounds most natural when you’re only shaving off a few decibels off the top. (Though sometimes you may want the effect of compression, in which case you can go hard.)

Limiting is kinda like when you take compression to the max. In the digital age, we’ve got brickwall limiters, which completely stop a signal from passing beyond a certain loudness threshold. Compressors do let signal pass beyond the set threshold, just not as far as otherwise. But brickwall limiters are like Gandalf. Brickwall limiters protect the Hobbits from the Balrog: that is, they keep a signal from becoming so loud that it “clips” (distorts) the output. Consequentially, the more you apply limiting, the louder your song can be. The Hobbits can safely advance!

Compressors and limiters are invaluable tools for making records. Most every record we listen to and love has been touched in some way by compression and limiting. But beware of too much of a good thing. Too much compression and limiting can kill the dynamics of your song (and even introduce other types of distortion, but that’s a whole ‘nother can of worms).

Dynamics are a big part of what makes music emotional. That crescendo swell into the chorus makes it hit so hard you want to start bopping along to it. A four-minute, four-chord loop may not ever get boring if it builds in intensity as the song goes on. If it’s all tamped down too much with compression and limiting, then what’s the point? If the Balrog weren’t a real threat, there’s no real story there.

So how do you know how loud you want your record to be? This is best discovered in conversation with your engineer or by experimentation. When working with a mixing or mastering engineer, they may ask for reference tracks: tracks you love the sound of, that match the vibe of your song in your imagination. One thing I do when I’m given reference tracks is purchase the CD quality files from Bandcamp or Qobuz and measure the loudness of each. That gives me an idea of my client’s loudness taste. When I’ve done and sent the first master, that’s usually when we talk loudness, in case the client wants it louder or softer in the next revision.

These days, with digital streaming being ubiquitous, most music is mastered no quieter than -14 LUFS. Many platforms normalize audio to about -14 LUFS, so as to avoid one song being much louder than another. Some argue that this might as well be the industry standard and that there’s no use in mastering louder since it will just be turned down by streaming anyways. I used to be a devotee of this philosophy until it lost me a mastering credit on an album. The artist asked if I could make it louder, but I resisted doing so, and I later found out the released album was mastered by someone else. This brings me to the consideration of vibe when it comes to mixing and mastering for loudness.

These days I’m of the mind that it doesn’t really matter if a record meets a specific target loudness, though I still have a preference. Many of my favorite records are around the -12 to -9 LUFS range. I don’t go quieter than -14 LUFS, unless I’m not mastering for streaming. These days, I mix and master more by vibe. Does this song need to be a banger all the way through? Or does it need to retain its soft-to-loud dynamics? The sweet spot is a matter of subtlety and individual taste.

As for me, I do still err on the quieter side as an act of resistance against the loudness war—the historic, ever-louder creep of commercial radio. I just think music sounds better if it’s a little more dynamic. Most of us are wanting to express something personal through our music, so as an engineer, I don’t want to limit the life out of it. It’s a balancing act to make something nicely loud.

Nothing is loud if everything is loud. This may be a maxim to live by in today’s attention-demanding world. What might it look like to release music in this age, when so many of us are either burnt out by loud advertisements or starved for actually engaging material or both? I like Jason’s saying. If it’s too quiet, listen closer.

And if you’re looking for someone to mix or master your record or someone to teach you how to do it yourself, you can visit my studio and lessons pages to learn more about what I offer. 🙂