The Recording Artist's Workflow: From Writing to Mastering
Everybody’s first few recordings – whether that means three or thirteen – usually go the same way. The deadline’s fast approaching and hardly anything is done. What has been done doesn’t sound right. Your fellow musicians have no idea what they’re supposed to be doing and nobody in the room is really sure what the next step in the process is.
For most beginners, developing a recording workflow isn’t a priority. But professionals will tell you that time is money and that’s why they’ve worked so hard on making theirs work. The quicker you can do your work, the more your effective hourly rate is, and the quicker you can do your work, the less you pay to hire studios, equipment and staff. It works from both an expense and an income perspective.
Here’s a look at the process and flow of a recording project that’s done properly, with tips for optimizing it further.
Writing is the first phase of any recording project, even if it begins well before the decision to record the song has even been made. During the writing phase the lyrics and the music are created and refined until they work well together.
Work Within Your Restrictions
When you’re in the writing phase it’s important to think about the realities of both performance and recording. If you write songs that you can’t feasibly record or perform live with your band, they’re not much good to you. How proud of your work will you be when you get into the studio and realize you don’t have the proper gear, musicians or instruments required to make the song work?
Building a repertoire is important and you need to be able to perform and record that repertoire. Even in the context of the recording process, live performance is important to consider. Unless you’re a studio-only band, your work needs to translate across all mediums.
Having a few restrictions to work around is conducive to creativity, so embrace the challenge. The artist who has every tool in the world at his disposal is often paralysed by choice, while working within a set of limitations can make creative decisions much easier. For most musicians, this is a recurring theme from writing to mastering, considering it can be one of the most expensive careers or hobbies around.
Prepare the Right Materials
Once you’re in the studio, you may need to work with session musicians who haven’t rehearsed the song with you, or you might want to discuss a certain section or riff in the song with your producer and fellow musicians. Backing vocalists will require lyrics. Failing to prepare the right materials in advance will turn out to be more of an impediment than many first think! Here’s what I recommend you bring to each session, at least a few copies of each:
- Tabs or notation for each instrument
- Lyric sheets
- Chord charts with lyrics
Although it may sound silly, rehearsal is part of the recording process. Because if you can’t play the song as a group in the first place, there’s absolutely no point going into the studio.
While having multi-track recording, comping and quantizing at the ready may make it seem like recording a song is easier than playing it live, the truth is that recorded music is really quite fickle and anything with more than two or three tracks requires a very, very delicate balance in the mix to prevent things from falling apart.
If you don’t sound good playing together, if you can’t lock into the groove of your rhythm section and make room for each instrument where each instrument needs it, it’s one more thing that will send the delicate balance crashing. Your recording will suck before it’s finished.
You need to rehearse until the song sounds polished enough that you think you could set up a few microphones in your rehearsal room and get a take you could be happy with.
While many a beginner may be tempted to shrug that advice off, here’s a more practical consideration for you: studios cost money. Every hour you don’t rehearse is another hour you’ll be paying the studio owner because there are so many problems with the recording.
Directing Focus in Rehearsal Sections
The most common method of practicing a song is to have the band play it repeatedly from start to finish. This is a particularly inefficient way of working, because you’re spending just as much time on the parts of the song the band is okay with as those parts that the band is struggling with.
Focus on different segments of the song, or even spend twenty minutes on a four-bar section of the song if it’s tricky for everyone to stay in step with each other. Don’t move on until the band has learned to play the troublesome segments perfectly.
It’s important to practice the song as a whole without stopping, but you’ll be spending your time much more wisely if you save the straight-through practices until everyone’s happy with the component sections on their own.
So, the song’s been written. The band can play it with their eyes closed, hanging upside down in their underwear. Now, they’ve decided it’s time to record it.
“What are we waiting for? Let’s record!” you all say in your excitement. Sorry to put a damper on your party, but not so fast. First you need to make a plan.
What will you do and in which order? How many instruments will be involved? How many parts will each instrument have, and thus, how many tracks for each instrument will be needed? How long will the whole process take? When do we need to book the studio and producer? The mastering engineer? The ever-elusive master piccolo player?
Without a plan, you’ve got two days of studio time left before the mix has to go for mastering and the only piccolo player still alive on this side of the equator is booked out for three months straight. Oops.
How to Make a Plan
A plan should always be in written form. I don’t necessarily mean pen and paper — it can be a computer document or a rock carving — but plans that are stored in the head of a human being are plans that are forgotten as soon as the planning phase ends.
Your plan need only be as complex or simple as the project at hand. If you’re in need of an outline to get you started, I’d suggest something loosely like this:
- End Product — clearly express what it is you intend to finish the recording process with
- Project — this is a comprehensive breakdown of the song and its components. Structure (verse-chorus-verse-chorus-bridge-chorus or what have you), some form of notated music — whether it’s guitar tab, a chord chart or music notation doesn’t matter — and a run-down of every instrument in the song and each instrument’s parts is important.
- People — list the people involved in the project and what their roles are. Think of the people you’ll need involved to realize everything in the Project section.
- Equipment — list the equipment you will need in order to realize what you’ve described in the Project section.
- Timeline — create a timeline for the recording process. The timeline should be comprehensive so you know when you start demoing, when you’re recording Moroccan percussion, which day you need to pick up the particular rack unit you’ve rented because it gives the song an essential sound, and so on. The purpose of the timeline isn’t just so you don’t miss sessions — it’s so you don’t miss anything involved with the project at all.
Plans can be a lot more complicated, so add to that as you wish. Plans can also be quite simple — if you want to record a three minute folk song through a standard rack of studio gear, with just yourself on vocals and guitar, and maybe a percussion overdub or two, you won’t need the same kind of structure.
For many musicians, the demo phase is one of the first phases of a project, and that means they’re not completely prepared and ready to get into the job. The first few days are spent pottering around the studio deciding what to do and thinking about the various changes that could be made to the song “just for fun”. While it’s all very creative, it’s probably not the best time to be playing around just because you haven’t got a firm plan of attack prepared, whether you’re paying for studio time or not.
With this workflow, you’ve gotten all that experimentation and uncertainty out of your system already. You know what you need to do, so you don’t feel the necessity to play around as a result of not knowing what the next step is.
The demo phase is the part of the project where each musician records a rough take that is good enough to form a framework that the rest of the project is built around. Once the demo is done and you move on to tracking, all you need to do is swap demo tracks out for the final tracks. The demo isn’t a polished, final track, but it’s important to get certain factors right. Production isn’t one of them, but timing sure is.
Perform to a Click Track
I really don’t care what your personal opinion regarding click tracks is — some elitists shun their use for some baseless reason or another — but when you record a demo, ensure everyone records to a click track.
Click tracks are hard to get used to for first-timers, so give each instrumentalist — in particular your drummer — some time to get used to playing with one. Once they’re used to it, be ruthless. Throw out tracks that don’t line up perfectly.
The reason being is that bad timing in the demo phase, since the demo will serve as your tracking framework, can ruin an entire project and weeks of work. One small rhythm screw-up can be easily fixed while you’re recording the demo. Halfway through the tracking process itself, it’s a big problem because you’ve got to redo an entire project that has been built on top of that error. The less vigilant you are and the longer you leave the problem in your project, the bigger that problem will become.
It’s also important to be picky. There’s no such thing as “not too big of a deal” because something that seems like a minor mistake will sound like a big mistake when every other instrument has followed its lead.
Short version: if every track of your demo is not in perfect time, the demo cannot serve as the backbone of your final piece.
Have Your Vocalist Sing to the Demo-in-Progress
Pitch is not as much of an issue as timing is when it comes to a demo, though you should ensure instruments are tuned well before recording them and do your best to achieve the best outcome here.
However, a slightly off-pitch instrument can throw off the vocal performance in the demo and/or the tracking phase. As the vocal performance is the focus of the public’s attention, you need to get it right.
Every time you finish tracking a particular instrument during the demo phase, let the lead singer sing along to it and ensure that there’s nothing throwing him or her off pitch. The problem itself may not be caused by out-of-tune instruments — it could be anything, including a timing problem — but whatever it is should be dealt with while you’re still in the more flexible and malleable demo phase.
At the tracking stage of the project, you’re using the demo as a guideline while you replace each track with the proper take.
Write Down Your Settings
Draw diagrams of rack mount equipment, take screenshots of plug-ins, do whatever you need to do to have a copy of your settings on file. This also goes for microphone placement: take a picture or draw one, just make sure you know how to put the microphones back the way they were.
And that means to the centimetre, so get out the ruler. One centimetre’s difference doesn’t look like anything to us, but from a sound perspective it can be huge.
You need to be pedantic about record-keeping so that if you need to re-track a section or you can only finish half of the tracking for a particular instrument in one day, you can pick up where you left off. Without an identical sound, you’ll generally need to start tracking from scratch. It might work as an effect here or there, but often having the tone of the instruments change halfway through the song is going to throw your listener off.
It’s Better to Go Nuts than Not Have Enough
So you’re not sure if the U87 or Classic II is a better mic for a part, and you just can’t decide. Record both, and choose later! It’s better to have options than to be stuck with a result you’re not satisfied with. The popular saying in the music industry is you can’t polish a turd. Generally it’s in reference to performance, but trust me when I say that the production values of a track can make an excellent performance sound like the proverbial unpolishable crap.
However, something you might not be able to change your mind about later is microphone positioning (you may well have seventy-three SM57s, but very few readers would have more than one U87 if they even have one at all). Spend time on positioning. Get it right.
Subtlety is Key to Destructive Effects
If you’re going to use outboard effects such as EQ and compression, be very careful and dial in the settings as subtly as possible. Use only as much of the effect as you need. Outboard effects generally sound better than plug-ins if you have great equipment or you’re in a professional studio, but you can’t go back and change things later. If you’re a beginner to intermediate engineer, you’ll often find yourself stuck with a track that you can’t use after you realize it’s ruined one or two weeks later. It even happens to experienced engineers often.
As I said regarding mic placement, spend time on your outboard (or otherwise destructive) effects. Get it right. And even then, make them subtle; if you haven’t compressed enough, you can fix things with a plug-in. If you’ve compressed too much, you’re screwed.
Routing often doesn’t get a phase of its own, and it may well not need one for many readers. If your genre is minimalist folk that will only use two to five tracks per session, you can safely skip this. Most electronica produces probably don’t need to make a big fuss over this. However, if you’re recording heaps of real-world instruments and have large, unwieldy productions sitting in Pro Tools or Logic or what-have-you, routing is an art form on its own.
There are two parts to my routing phase:
- Setting up instrument group auxiliaries
- Setting up send/return systems for common effects
If you’re not sure how to set up a group aux sub-mix or a send/return effect, there are a few tutorials on Audiotuts+ that explain each.
If you have a song with 30 or 100 tracks in the project, it’s too much for your mind to handle in terms of relative mixing. Break everything down into a few groups (less than ten, preferably less than five) such as vocals, drums and guitars.
Then re-route each of these tracks so they go to a stereo auxiliary channel rather than straight to the stereo master. Get a relative mix between all tracks that are being routed to the auxiliary, and spend some time on it to make it perfect. If the sub-mix is good to go, you may not have to touch individual track levels again; just move one of your group aux faders to adjust the mix.
Mixing as few as 20 tracks and as many as a few hundred is a painful process. If you can do that many using between three and ten sub-mixes, you’re laughing.
There are commonly used effects that are used in almost every production: reverb and delay are the most common, and chorus makes an appearance in most projects. You should set up send/return systems so each track sends a certain amount of its signal to a single processor that caters to every track in the project. This has two benefits.
- Using a reverb (or other effect) plug-in for every track will kill the fastest of processors
- Using a reverb (or other effect) plug-in for every track will result in a sloppy, non-cohesive sound
It’s very easy to ruin a song with the abuse of effects. While it can still happen with the use of send/return systems, it’s much easier to control. Different reverb settings on different instruments sometimes works, but often it will confuse the listener’s mind. Using effects with the same settings, through the same processors, helps tie the mix together and make for a more cohesive sound.
Before you get down to the mix, you want to ensure that all your audio is in place, lined up, cross-faded and generally ready to be mixed without requiring any big changes or presenting any structural problems. You need to be able to focus on the levels and effects used in the next phase without distraction.
It helps to have an editing checklist prepared that you can use every time you record something new. You might use something like this, although it’s by no means complete:
- Edit Regions
- Fades & Crossfades
Editing regions involves cutting up various takes and lining up the best comp track you can. 95% of editing time is usually spent here, and that’s because of the drum kit. You’ll need to go through each drum track and ensure that the transients line up with the beat. That can take hours to do right. You’ve got to zoom in pretty close as well, which slows things down — this prevents you from snapping things in line too mechanically as would happen with a normal zoom level. At a normal zoom level, any sounds that you move will travel further to snap to grid, creating a technically perfect but emotionally void performance.
Once every region is in place you need to create miniscule fades on each region to prevent popping noises all over the place, and ensure that every splice between takes is cross-faded.
Once you’re satisfied, many people consider it a good practice to consolidate regions so working with them from here on out is easier.
Finally, while anything to do with changing levels is generally considered part of the mixing phase, it can be a good idea to use automation to even out the volume of wildly variable tracks before you get to the mix. Mixing the first half of the verse only to find out the second half of the verse has totally different dynamic characteristics is not fun and can require a lot of work to fix (the easy solution is to whack a really tight compressor on everything, but if you want to suck the life from the sound go right ahead!).
Mixing is such a simple process in theory and yet can become so complicated and stressful in reality. The idea is to get each sound controllable and aesthetically pleasing with effects and processors, and get the levels right. My process goes something like this:
- Relative levels
- Solo a sub-mix and re-do the relative levels on each
- Then mix the sub-mixes until the song sounds balanced
- Processing & Effects
- Work on the basics for each track until I’m perfectly happy with the sound: EQ, compression, limiting. Guitar amp simulators are one of the “basics” if I’ve recorded electric guitar as DI.
- Only once those are done will I move on to other effects that are more about changing the sound and less about controlling and improving it.
- Final levels
- Processors change the sound and often the levels of each track. Go through your sub-mixes and your overall mix again.
- Rinse and repeat this process until you’re happy.
Mixing is really quite a big job, so without some sort of plan — even one as basic as the one I just outlined — you won’t know where to start, or when you’re finished. I know that I end up doing a lot more than I listed, often in a different order than the one I listed. But if you don’t know where to begin or how to approach the job, your mixing session will take a lot longer and you’ll be confused by problems with everything from your gain structure to trying to remember what you’ve processed and what you haven’t.
This is a fairly overlooked part of the post-production process. Once you’ve got a final mix, you want to create a few different versions of it and bounce them to disk. Each version will feature a mild change that represents a small change in focus on the instruments in the song. One with a 1dB louder lead vocal, one with a 1dB quieter lead vocal, one with more pronounced backing vocals, perhaps one with quieter drums, and the list goes on. Usually changes won’t be more than 1 or 2dB.
You do this because mastering changes the sound of the mix and this gives you the ability to quickly listen to various mastered versions without going back, changing something, bouncing it, mastering it, and then comparing it. Think about the most likely changes you’ll want to make and get them all ready and bounced so that they can be mastered in one go.
If you’re not doing the mastering, which is probably the best idea (unless you’re a mastering engineer with a mastering studio), you should still provide the mastering engineer with these options. They will often call you up and ask for them if you don’t provide them to begin with. Thanks to Audiotuts+ Plus contributor Bobby Owsinski for this tip.
I don’t pretend to know much about mastering — my workflow is to get my multiple versions ready and pass them on to someone who does pretend to know. The best advice I can give you is to send your music on to mastering engineer. When I need to do it myself, it’s not a particularly scientific method at all; I bounce songs and important them to an album mastering session, apply the usual processors, and try to get the volumes of each track aligned as best as I can.
That said, it’s important that mastering is part of your plan, because you’ll need to find a mastering engineer, book some of his time, and remember to send your work over to him.
If you do insist on mastering yourself, be sure to check out the Audiotuts+ website, as it has some useful tutorials on the subject. If you never intend to use a mastering engineer, do yourself a favor and get a more advanced education in the subject.