By Rob Stewart - JustMastering.com - June 21, 2012

Last updated June 28, 2015


What goes into mastering a song? What are you paying for when you hire a mastering engineer? There are many books written about the art and science of music mastering and most of them are excellent resources. Rather than go deep into the science and theory of audio mastering, I will attempt to provide a high level description of the mastering process applied here at JustMastering.com.


Mastering a song

Depending on the song, and what is required, the entire process can take anywhere from one to four hours, and much of that time can often be consultation time with the producer and mix engineer. Time spent on audio mastering can be divided into four steps that I generally follow when mastering any song, EP or album project.


 

STEP 1 - Critical Listening

It is true that a good audio mastering engineer can greatly enhance the sound of your recordings and mixes, but to do this, we need to start with the best possible foundation. Mastering cannot add what doesn't already exist, so if a mix is lacking fidelity, clarity, impact, expression or drama, it is best to go as far back as possible to either the original recordings or the mix to find out where the issue started, and what production steps are necessary to improve the mix. 


Bearing the above in mind, I spend a fair amount of time listening to and analyzing each song I master. For example, what type of music is it? How well has it been recorded and mixed? Are there any serious issues with it that should be addressed before mastering? Taking this time to understand exactly what I am working with is one of the most critical steps of the mastering process. In my view, mastering music is not about rushing to process a mix file, or starting with a standard "cookie cutter" approach, it is about assessing whether or not the producer and mix engineer met their goals for the song (and if not, what can/could be done at the mix stage), and if there are other technical concerns that need to be addressed as well.


My role at this step in the process is to help my client achieve the best mix possible, before we proceed with mastering. Distortion and over-compression in a mix are the two things that are most challenging to deal with from a mastering perspective. Distortion is broadband noise, so it cannot be removed with EQ. It can sometimes be reduced but it almost always seriously impacts the tonality of the recording. If the original track is distorted (cannot be fixed in the mix), then I would even suggest re-recording the track.


Oct 15, 2013 Update - I have since started offering a detailed mix analysis report as part of my mastering services. I created this report primarily to help provide all of my clients with consistent, personalized practical advice to help them improve their mixes before we move on to the final mastering stage. 

STEP 2 - Diagnosis

Based on my analysis, and after any mix adjustments are made, I determine what else the song needs, based on the client's goals. Assuming that we are starting with a mix that has good structure, balance, movement and tonality to begin with, we'll now start to focus on more minor issues that need to be dealt with. Some common issues I hear a lot are issues with too little or too much bottom octave/sub bass, and too little or too much in the top end. It is normal for any mix to have to little or too much in one octave or another, for various reasons. Whether that type of issue is addressed via mastering, or by revisiting the mix depends upon my client's time and funding.  


Diagnosis is probably the trickiest part of the song mastering process. I believe it is in our nature to want to apply as many processes and effects to our music as we can. It's so important to step back, listen and ask "what processing does this song need"? It is a clich√© these days, but less is definitely more both in mixing and mastering a track. I only use processing that I believe is required for the song, and every song is different.



STEP 3 - Processing

Based on my diagnosis, I apply the required processing to the song. Mastering engineers use many of the same tools as a mix engineer - equalization (EQ) and compression for example. The difference is in how we use those tools.


For example, a mix engineer will use EQ to carve some space in their mix for an individual track or to remove some background noise/rumble from the recording. They might use a high pass filter on a guitar track to remove extreme bass, or they might use other forms of EQ to bring out the edginess of a distorted guitar or the breathiness of a vocal track. Mastering engineers use EQ to make much gentler and broader adjustments to the mix as a whole. Maybe the mix needs less bottom end? Maybe it needs more air or even a lift in the upper mids? Most often, the EQ adjustments are subtle, and intended to help balance the sound of the mix as opposed to altering the character of the mix in a material way.


The same applies for compression. A mix engineer will sometimes apply very heavy compression to a vocal track to give it an "in your face" sound, whereas a mastering engineer would typically apply much lower levels of compression to the track if at all. In fact, I typically work within the ratios of 1.5:1 to 3:1 if at all.


Limiting is another mastering tool that helps catch any extraneous peaks (i.e. not to maximize loudness - more about loudness, here). I generally apply very light mastering limiting, again depending on the song, and the needs.


There are occasions where I will use harmonics enhancement if I feel the song calls for it, but it is rare. It depends heavily on the recording and the mix itself (many mix engineers are already adding harmonics and saturation to tracks within the mix already). The article "My Audio Mastering Philosophy" contains some more information about this, and my general approach. 



STEP 4 - Finishing

After all of the processing has been applied, I will apply fade-ins/fade outs as necessary, and will check the track's gain level. I routinely involve my clients in this step because everyone has different goals, and the gain level of a track will depend greatly upon how my client intends to publish their song.  


I like to set tone, loudness and gain levels bearing the music and song sequence in mind.  For instance, a ballad should generally not sound loud or aggressive (though I'm sure there's an exception somewhere - grin). Note that any song can be too loud. I educate my clients about this every chance I get. You can achieve a high "perceived" loudness with high amounts of compression and certain EQ choices, but the trade off is that you lose musicality, and the resulting sound can be very fatiguing. No one wants to listen to music that is hard on their ears, so I encourage my clients not to be too concerned with how "loud" their project sounds compared to others. It's important to focus instead on how good it sounds.


There are many examples of commercially mastered music on the market today that have been over-compressed in an effort to make them as loud as possible. I encourage my clients not to follow this path. It's much more important to emphasize the musicality of your song, and to bring out all of the energy and dynamics of the music. That ultimately will make the mix sound bigger and louder when it needs to. You can read more about this topic in my loudness article


At the very end, I trim the heads and tails (start and end) of each track and add any requested metadata (including ISRCs) to the files.



Mastering an album or release

Most clients who are looking for mastering have a full album or release. The process for an album is similar to what I've described above, except that with a release, I also pay close attention to the gain levels of each song, in relation to each of the other songs on the album. Please note that it is not about making each song the same loudness level as the others, it is about making each song the appropriate gain level in the context of the whole album.


There is a timing component as well. The lead in and lead out for each track is timed musically so that there is a nice flow from song to song.


Tonality is important, too. Some mixes are brighter than others. Maybe they were recorded in different places, mixed by different engineers etc. A big part of mastering a full album release is to help the whole release sound cohesive. That does not mean making each mix sound the same, it is more about balance. I will sometimes suggest alternate sequences to my client at this stage for tonal, musical, or mood reasons where I believe a different song order might be best.


I sincerely hope that this helps explain the mastering process. If you have any questions please drop me a line any time.

Looking for feedback on your mixes?