What You’ll Learn

  • Where did loudness normalization come from?
  • When do differences in loudness get noticed?
  • Does loudness matter when submitting your song to a label?
  • What happens ACTUALLY happens when A&R listens to music
  • How much loudness difference can you notice?
  • Why you should submit a competitively loud master to labels
  • Does loudness matter for purchasers of your music & DJs?
  • Should you master to -14 LUFS for Spotify and Apple Music? 
  • Loudness targets…do they matter?
  • What do other mix & mastering engineers think? 
  • Why most people can’t mix or master loud
  • How to create the potential for loudness
  • Don’t be scared of loudness

All About Loudness

Loudness Trumps Everything is an article in the Journal of the Audio Engineering Society from 2004.  And this was the starting point for the discussion around loudness normalization standards in the industry.  To condense what it says, it’s basically stating concern primarily for consumer playback experience in the era where CD players allowed you to put in multiple discs and shuffle them such that songs of dramatically different loudness could now be head back to back, resulting in jarring loudness jumps.

Today’s equivalent is the Spotify playlist or the music queue.  Why do I bring this up?  Because loudness differences matter MOST when songs are being listened to back to back, in immediate proximity to each other.  Auditory memory in humans is very short – a matter of seconds – so it’s only this fast back to back listening experience where loudness jumps are overtly noticeable.  I want you to remember this fact throughout this video.

There are a ton of videos on loudness, some advocating for it, and many advocating against it, when mixing and mastering.  Usually the ones against loudness cite similar points like: 1. Streaming sites like Spotify turn things down anyways, 2. Masters with a lot of crest factor sound better, 3. Or DJs can just turn up the gain if a track is quieter than the rest of the songs in their set.

I’m not going to get involved in that side of the loudness argument as it’s been discussed ad nauseum.  What I WILL do in this video is cover new ground by presenting 3 essential things to consider about loudness that I have not heard discussed yet.

Why listen to me?  I’m a mixing & mastering engineer (whoooo who isn’t these days) – but I’m working professionally, earning an income, working with multiple clients.  I have a business degree and I’ve worked extensively in finance; I have an appreciation for how money works in the industry and what it takes to generate financial sustainability.  I run A&R at Warp Academy vetting thousands of submissions song for production contests, hiring producers and sound designers, and I sign songs for our label.  So I’m very aware of the psychoacoustic aspects of loudness and what it does to the perception of quality and the establishment of subjective preferences. 

Context is essential here: I’m talking about modern pop, hip hop, and electronic dance music intended for club and festival play by DJs.  That’s our container for this conversation.  Let’s get into it. 

So imagine that you’ve just finished a song and now you want to have it released.  You want the prestige of releasing it on a prominent label so you start shopping it around.  How do you send it out?  You either upload it to a private streaming player like Soundcloud, or perhaps the label gives you an upload form.  The biggest labels, like Monstercat for example, only let you select 20-30 seconds of your song to submit.

That’s ALL you get.  20 seconds.  I talk to other folks in A&R all the time.  Even if you send a full song, all you really get is 20 seconds.  When we’re listening to hundreds of songs a week, all you get is a moment of our attention.  We don’t listen to the full song.  We don’t adjust the volume.  We just skip to the drop and play it.  That’s usually how it goes down. 

So what did we establish earlier in the video?  It’s rapid back to back play where differences in loudness matter the most.  Is there any loudness normalization for A&R people listening to submissions?  No.

Here are some other facts: It’s been proven that listeners will be biased towards music that is louder.  How much louder?  As little as a 0.2 dB difference in loudness is enough to create a preference. Certainly 1 dB can.  So if your song is not competitively loud with the other submissions, or EVEN the song played right before yours, you’re damaging your changes of getting signed to that label.  Once the gatekeeper forms a first impression of your music, it’s very difficult to change that.

So this first point is about getting your music signed to a label.  If you can’t do that because your song doesn’t stand up, you’re dead in the water. 

Okay, so let’s say you’ve passed the first hurdle and you get your song signed to the label.  What matters to you most now?  Streams on Spotify?  Sure yeah.  But what likely matters more is that DJs buy your music and play it in their DJ sets.  How do DJs listen to your music?  They listen to tiny clips of songs in rapid succession back to back on sites like Beatport with NO LOUDNESS normalization.  Are you getting the point now? 

Little differences in loudness between song previews will be a major deciding factor in the buying decision of that DJ (because…psychoacoustics).  Even after that, the DJ is going to be browsing songs as they assemble their set, once again creating the perfect storm of conditions where loudness matters: fast back to back playback of tiny snippets of songs.  

Now we’re going to go after the Spotify thing.  Loudness normalization in Spotify is an end user preference.  Same in Apple Music.  All music is delivered to the app at full original loudness.  Users can selectively turn the normalize loudness feature On or Off.  You have no idea if the user has this enabled, so mastering your song to -14 LUFS ensures that everyone with it turned off is going to hear your super quiet master.

Also -14 LUFS just happens to be the loudness target that’s popular today.  What happens if that changes in 2 years and they decide it’s -10 now.  What do you think happens to all those -14 LUFS masters?  They aren’t going to turn those up most likely.  They’re going to turn everything else down to that level.  This is EXACTLY what Youtube does.  It turns things down to -14 LUFS, but will not turn quieter things up.  

Here’s what another prominent mastering engineer who I’ve done lot of work with has to say:

There was a brief period a couple years

There was a brief period a couple years ago where everyone wanted streaming specific masters, but that seems to have died out. Too many issues with songs still ending up playing back quieter than people intended, you just can’t account for how many people will have volume normalization on or off on their end. These days we’re back to just making one master at the typical loudness levels and using that for all outlets. At least that’s how it’s been for say 97% of my clients.

Now, I am NOT advocating everyone pushes their songs super loud.  Most producers and mixers don’t actually know the techniques necessary to do this, and as a result they end up destroying the song in their attempts.  That’s what I call GARBAGE LOUDNESS and it’s very common.  That said, it is completely possible to create very loud masters that are clean and sound great.  But you need to know the secrets and techniques and tools.  

The foundations for loudness has very little to do with the mastering phase.  I starts from the very first sound you choose in your song, it has everything to do with how you arrange sounds on your timeline, and the potential for loudness in mastering comes in the mixing phase.  You have to nail all of those things otherwise, you get garbage loudness.  But you shouldn’t be scared of loudness either. 

Like what you see? sharing is caring!