What Is Mastering And Why Does It Matter?

I think mastering is the most misunderstood part of the process in the recording industry. Until a few years ago, for all I knew, mastering involved taking a song into a dark room, muttering ancient spells over it, and declaring that it was indeed finished. In this post, I’m going to give my perspective on mastering, and some helpful information about it that will hopefully demystify mastering for you, and better evaluate:

1. Is my song mastered and ready for release?

2. Was my song mastered well?

What Is Mastering?

Mastering is the “last” step in the process of making a recorded song. Once a song is mastered, and the master is approved, the song should be thought of as finished. I’ve heard it said that on the American side of the Atlantic, mastering is thought of as the last step in the engineering process, and on the European side of the Atlantic, it’s more usually thought of as the first step in the manufacturing process.

Mastering is primarily quality control. It’s the process of listening to a final mix with fresh ears, in an acoustically well-treated environment, and determining if the mix will “translate” on whatever mediums that mix will be distributed to (vinyl, tape, CD, streaming, radio, synchronization).

If the mastering engineer determines that there are subtle ways that the final mix could sound better, they will make those changes in hopes of making that final mix translate better.

In a perfect world, the mastering engineer would make no changes, and determine the final mix is perfect and ready for distribution.

As the music industry has changed, and the technology that we use to record and listen to music has changed, mastering has changed.

In order to understand the current landscape and role of mastering, it’s important to understand the history of mastering and how we got here.

How Did Mastering Start?

Mastering started when all recording was done to tape, and then that tape had to be sent to lathe cutting where a vinyl cutting engineer would use a lathe with a diamond or sapphire stylus to manually cut grooves into the vinyl “master” disc (why this process was called mastering). The master disc is then used to create metal stampers that are then used to create the distributed vinyl records for consumers to purchase.

There are physical limitations that tape and vinyl have, and sonic characteristics that are built-in to recording to tape and mastering to vinyl. If a mix on a reel of tape has too much low end, harsh sibilance (high end), or is too loud, and the lathe cutting engineer tries to cut the vinyl without any changes, the cutting process can “blow a head” permanently destroying the lathe cutting machine, or create a vinyl record that is unplayable by a record player (the needle will skip).

The lathe cutting engineer was supposed to cut the master as faithfully to the final mix of the tape as possible. The preference was for the mix to be great and the lathe cutter to not have to do anything to the mix during the cutting process.

But, there were and are times where the lathe cutting engineer has to apply some (hopefully) subtle changes to the overall EQ (the balance of low and high end) and the overall loudness of the mix in order to successfully cut a clear, usable, great-sounding master.

Obviously, vinyl is still being made and listened to, and it is largely made the same way. Lots of music that is made into vinyl today isn’t recorded to tape, so in those cases, the final WAV or AIF file of the mix has to be turned into an analog signal (similar to a tape) that is sent to a lathe cutting machine.

Tapes, CDs, mp3s, and now streaming all have different sonic limitations and characteristics that impact what sonic characteristics can be distributed on them.

Does My Song Need Mastering Before It’s Released?

Before you release/distribute your song on a given medium, it’s best to have someone with experience mastering songs for that medium to master your song.

Traditionally, the person mastering the song is a different person than the one who recorded and mixed the song. Increasingly, with tighter budgets and music professionals having to become more diverse in their knowledge and skillsets, people are recording, mixing, and mastering the songs they work on.

If you don’t have much experience mixing, or aren’t mixing very loud, and decide to send your mix to a mastering engineer, there’s a high probability that you will receive a master that sounds very different than your final mix. Some of those differences will probably be great, some may not be. This can be really disconcerting, and has more to do with your ability to mix, than the mastering engineer not knowing what they are doing.

What is “Loudness”?

Before I start getting into a bunch of details about how loud a master should be, let me go over what loudness is and how it’s measured.

The loudness of a mix or master is measured in LUFS, which stands for “Loudness Units Full Scale” or “Loudness Units Relative to Full Scale.” LUFS measures loudness over time. This is different from measuring decibels, which measures momentary loudness.

Decibels measure how loud a sound gets at its peak. Most natural, uncompressed sounds have a really sharp peak. This means that the moment when they are loudest is extremely short. If you go and clap in your living room, you might hear some echo 0.5 seconds after the clap, but most of the noise will happen in those first few milliseconds.

If you compare that to an organ, the loudest point of a note could last for minutes, or theoretically as long as you wanted it to.

Things like vocals, drums, acoustic guitars, and even electric guitars, basses, pianos, and the majority of instruments in your average song, have a pretty sharp attack and a pretty sudden dip in volume.

Here’s a waveform of a vocal recording. You can see it’s got really drastic changes in how loud and soft it is. The spikes are the loudest spots, and the valleys are the quietest. If this was a mountain range, hiking it would be impossible. There’s going to be brief moments when this vocal is going to feel piercing, and syllables of words that will be pretty unintelligible when music is playing underneath it.

Sounds that have these really drastic shifts in loudness can be described as “dynamic” meaning: they achieve very loud peaks, and very quiet valleys. When you listen to super dynamic sounds, your brain focuses most of its attention on the really loud sounds, and doesn’t really notice what’s happening in the valleys.

This second wave form is the exact same vocal recording, but it has been compressed, meaning that some of those sharp peaks have been pulled down, and some of those quiet notes have been pulled up in volume. Even though the second wave form doesn’t reach the same peak decibel level as the first, the average loudness across the performance, or loudness over time (LUFS), is higher than the first, and your brain will perceive it as louder, and be able to discern more of the details of the overall performance than the first uncompressed wave form.

Compression is used throughout the recording and mixing process to make instruments more consistent and interesting for the listener.

How Loud Should My Master Be?

It depends on your genre and your goals as an artist for your song and/or album.

Some genres like pop, heavy modern metal/rock, EDM, and hip hop use lots of compression, and have really high LUFS values. Typically, songs in these genres are trying to sound loud, exciting, in-your-face, punchy, driving, and energetic.

New releases of these genres can easily sit at -6 LUFS or even -4 LUFS at the loudest parts of their songs (the closer the number is to 0, the louder it is).

In general, over time, music has gotten louder and more compressed, so genres that are younger are going to be more compressed and louder.

Genres like jazz, blues, bluegrass, classic soul, classical, folk, and even classic rock, are going to have more hit songs that are mastered at lower loudness levels and rely less on heavy compression.

Jazz and classical music might sit around -25 to -12 LUFS.

With genre being so fluid these days, it’s really helpful to have artists in mind that you want your music to be associated with, and understand what your brand is. You may not want to feel tied down by genre, but you should have a clear understanding of what world your art lives in.

For example: hip hop frequently samples classical music, but that doesn’t mean it needs to be recorded, mixed, and mastered like classical music. Knowing what features, processes, and sonic characteristics your music should have will help you make effective decisions in the recording, mixing, and mastering processes.

Artists who are making genres like indie rock, alternative folk, indie pop, need to make sure that they aren’t underutilizing compression during recording and mixing, and then sending it to a mastering engineer who is trying to then use lots of compression to get the song(s) to a loudness level that’s competitive with pop artists, or masters will sound overcompressed and strange.

LUFS Are Not All Created Equal

The numbers aren’t all that matter. One song that is mastered at -4 LUFS isn’t necessarily going to sound the same as another song mastered at -4 LUFS. Getting a song to be loud without sounding distorted is an art that takes lots of skill, experience, and a great ear.

In general, the louder a song is, the more distortion and digital artifacting it will have. The quieter a song is, the easier it will be for it to sound clean and clear, but the less in your face it will feel.

A good mastering engineer will understand what loudness and tonality will be best for whatever song(s) they are working on, and achieve a level of loudness and compression approach with that in mind.

The Loudness Wars

“The loudness wars” refers to mastering engineers trying to master songs louder and louder than the “competition.” It’s built around the idea that most people perceive louder masters more positively.

The loudness wars started in the 40’s and 50’s, became a huge focus in the 90’s and 2000’s and still rages today.

Streaming Platform Recommendations

When streaming platforms like Spotify came around, they released recommended loudness levels. To this day, Spotify recommends that songs are mastered to -14 LUFS. The idea was that streaming platforms put lots of songs next to each other in playlists that are mastered at lots of different volumes and that Spotify would use their own limiting processing to “normalize” the loudness levels of songs, effectively ending the loudness wars.

At the birth of streaming, mastering engineers started trying to master songs to -14 LUFS, but it didn’t take long to realize that Spotify wasn’t necessarily going to make a song mastered at -14 LUFS sound the same as a song mastered at -4 LUFS. Even with Spotify’s limiting, louder masters will still usually sound louder than quiet masters.

If you are on a phone or desktop, you can actually turn on or off Spotify’s loudness normalization, and even apply EQ(!) to all the songs you listen to.

But, it is almost always not a good idea to shoot for that number (-14 LUFS). Most songs are going to have a Loudness sweet spot that’s dependent on how dense of an arrangement they are, what instruments they feature, what genre they are in, and how much compression they used in the mix.

Most modern songs should clock in somewhere between -10 LUFS and -4 LUFS in my personal opinion.

Mastering Is Limited In Scope (pun intended)

In the mixing phase, the mix engineer is working with every track that was recorded to make a great sounding mix. If you’re working with a dedicated mastering engineer, that engineer is working with a stereo file of the whole song.

Therefore, any changes they make will affect every element in the mix.

Mastering will not make you like your song, and usually won’t make a bad mix good. If you aren’t happy with your master, the chances are high that there was something wrong with your mix that your mastering engineer is trying to put a bandaid on, or that your mastering engineer is just doing too much.

Mastering can raise the overall loudness (LUFS) of your song through compressing, limiting, and clipping the mix, change the overall balance of the song through applying equalization to your mix, and widen your mix slightly.

Mastering Is Subjective

Like every other part of music creation, mastering isn’t a science, it’s an art. Every mastering engineer will have their own tools, processes, tastes, strengths, and weaknesses.

In this video, metal mixing engineer Jordan Valeriote shows how even the top mastering engineers can make mistakes…

https://www.youtube.com/watch?v=E9ssuVxXCdI

Don’t Use AI Mastering

Lander, Distrokid’s Mixea, and I’m sure hundreds of tools out there now are AI mastering tools that are supposed to help you get a polished final master. If you are serious about your music at all, I wouldn’t recommend using any of them.

In my experience, all these tools do is make your song louder, brighter, and wider. To me, this is missing the point of what mastering is supposed to do.

You don’t finalize a mix and then say: “ok now let’s juice it up a little.” Some mixes will be too thin, some will be too dark, some will be a little too mid-rangey, some too scooped. Some will be perfect just the way they are. It’s best to send your song to a human and get real human feedback on it, and be able to ask them what they think needs to change and why.

If you’re serious about your music, and you’re not confident that your song sounds finished, get a real person to give you some feedback. *** If you used Distrokid to distribute your music, it will automatically try and master your song for you using their AI mastering tool called "Mixea." In pretty much every case, you should opt out of this option. It can be easy to skip and accept Mixea's mastered version for your release if you aren't careful.

Albums Vs. Singles

Releasing albums has become less common these days, but I think that an album can create some helpful parameters for mixing and mastering.

When you work on recording, mixing, and mastering an album, a mastering engineer will have more context for what you want to sound like and what characteristics fit your music as a whole. In an album, you don’t want songs to sound disjointed. In an album context, mastering takes on a greater role, because it can be difficult to impossible, to mix every song in an album to the same loudness and to complementary tonalities (some songs will be a little bright, some might have a little too much low end, etc.)

A mastering engineer, (and a mixing engineer if they get all the songs at once) can use every song on the album to inform their decisions. They might be tempted to slam one song on the album with lots of compression, but then hear another song on the album that won’t work at all with that approach, and find an approach that makes every song work together well as a cohesive piece of music.

In the end, I think the album approach has a higher likelihood of every song getting mastered in a way that creates a cohesive sound, and establishes a strong brand as an artist.

The Song Is Done When You Say It’s Done

I find Steve Albini’s perspective as expressed in this video helpful:

https://youtu.be/TQonl6exhNU?si=7e8QJsXsNc4b-uuh&t=2459

The struggle of every artist in every artistic discipline is learning when to walk away from their work. If we aren’t careful, mastering can be the equivalent of tapping out of the music creation process. Make sure you listen to your master and that it sounds exactly how you want it to sound.

Sometimes you’ve just got to walk away from the song and call it good, sometimes you need to revisit the mix, sometimes you realize you actually didn’t like the song and don’t need to release it.

Don’t beat yourself up about it when one of these things happens, allow yourself the freedom to have an opinion about your music and go with your gut.

My Approach To Mastering

I don’t consider myself a specialized mastering engineer, but I do master a lot of the songs I mix these days, and do master other people's mixes from time to time. When I first started mixing, I kept having mixes approved, sent to a mastering engineer, and then sounding different than what the artist approved. Sometimes the master didn’t sound as good to the artist as the mix. Most of the time, it was because I had mixed the song really quietly, and when the mastering engineer made it louder, we could hear things we didn’t like.

I realized, I was treating mastering like a scary, mysterious thing, and that my ignorance was compromising my ability to provide artists with great, finished sounding songs. So, I decided to offer artists a master of their mixes, and if they didn’t like it, we could send it to a mastering engineer.

Learning about mastering helped me pay more attention to arrangement, compression levels in recording and mixing more, and helped me get a better feel for how songs feel at different loudness levels.

I try to give my ears a break between recording and mixing, and between mixing and mastering. I always let artists know that if they aren’t happy with how the song sounds, that we can pull in a dedicated mastering engineer.

Most of the time, the artist is happy with my final master.

Do you need a song or album recorded, mixed, or mastered? Reach out to us via our “Get Started” form and we’d love to hear about your project.