Tarekith's Guide to Song Arranging
One of the most common issues I see new producers having, is turning that 8 bar loop they made into a full-fledged song. You've spent hours crafting a funky groove, you can't stop bobbing your head to it, and all your friends assure you that it's easily the greatest sounding bit of audio (all 2.3 seconds of it) they've ever heard. So, how do you take this little bit of an idea, and turn it into something that people will want to listen to over and over again?
In this article I'll look at different ways I've found that speed up the process, and make it a lot less daunting than you might think it is. As usual, the same caveat from all my guides apply here. Namely that you should never take what I say as gospel, don't be afraid to try new ideas of your own, or tweak the ideas presented here to suit your own style of writing.
Before getting into the ways that you can turn a loop into a finished song, let's take a quick look at why it is that so many people get stuck in that "loop creation" phase in the first place. I think the main reasons are two-fold:
- The gear dictates it.
- The increasing popularity of pre-made loops.
Many people using hardware groove boxes know the concept of the "pattern", short musical phrases typically less than 32 bars long. These are then arranged into longer songs, which are basically nothing but collections of patterns. So from the get go you're already being trained to work on writing music that works as a shorter sections of repeating music. And it's not just hardware where this mindset is in use, software manufacturers have mimicked the way these devices work too, for good or bad. Programs like Reason, Fruity Loops, and Acid all have the pattern concept at their core. As these are typically the type of programs beginners are likely to purchase and use, it's easy to see why people have trouble getting out of this stage.
The other reason I think people get stuck in this loop mentality, is that more and more people are turning to pre-made audio loops in the song creation process. It's relatively easy to fire up a program like Ableton Live or Sony's Acid and start combining a bunch of loops into a good sounding groove. You can get something that sounds like a finished section of a song in no time, without really needing to think about things like song structure and keeping people's interest up over the length of the song.
So then, how do we take these elements and turn them into something we could call a finished product?
The first thing you need to do is think about what you're trying to say with the song, what is it's purpose? Is this something that will be a full on dance tune where you want the energy level up the entire time? Is this a song that you want to build gradually, or is it a closing piece on a CD that needs to slowly fade down in intensity? Or do you really not have an idea yet, just some cool sounds you want to make into a song? Either way, write down your intentions, and keep them handy. A lot of times it's easy to get distracted once you start, and being able to refer back to your goal while working can save you a lot of headaches later on.
The first thing I'd recommend is setting up your DAW so that each of the sounds is on it's own track in the arrange/project/timeline view. This lets you easily move the different parts of the song around to form the outline of it. It can also help you visualize the layout of things as well, which can be a big help when you're just starting out. It doesn't matter if you're working with midi or audio, the premise is the same in either scenario.
Decide on a length for the song next, at least a rough idea anyway, you can always change it later if you want. For most dance music, I'd say 7 minutes is a good starting point. Other types of music can probably get by with 5 minutes initially. Again you can always change this later so don't feel too constrained by your decision and stress over it. A lot of times people try and write these huge epic songs, and end up with a song that just doesn't progress very well, so don't feel the need to create a 12 minute journey all the time. Keep it short, keep it sweet, and you'll likely hold people's interest much better.
So now we have our blank canvas (the empty song project, zoomed out to show the whole song length), and we've got our song elements (the parts of our loop) spread out vertically. It should look something like this:
I think this is where a lot of people get stuck, or just confused on where to go next. It's sort of similar to writers looking at a blank page and not knowing where to start, or a painter looking at a blank canvas and being afraid of making that first brush stroke. So we're going to get past that stage right away, and just fill up our canvas with color as it were. Go ahead and copy all of your loops for the entire length of the song. Fill that canvas up with all your parts so you're not looking at a blank page anymore. You should be seeing something like:
This is a technique I call subtractive sequencing, where we start with all the song elements and remove bits and pieces to form the structure of the song. We'll come back to this, but first let's talk about song structure.
In it's simplest form, this is nothing more than how your song progresses from start to finish. What elements are playing at the same time, when do they each sound, how long do they play for, etc. It's how the energy of the song is laid out in relation to the song length, which is one reason it helps to have a good idea of how you want your song to be structured when starting to arrange it. But even if you don't have any ideas on this, there's some simple guidelines that can make it all much easier to figure out.
Take your loop for instance. In all likelihood, when you are playing all of the elements of the loop at the same time, this is the most energy your track will ever have. You're not adding anymore audio information since you're using all the parts, so it's pretty safe to say this is the 'peak' part of your track at this point. You can build up to this, or you can take away from this and reduce the energy, or you can build and release from this point many times throughout the song. It's really up to you, and this is one place I can't tell you what to do.
But for the sake of learning, let's say we want to create a tune that gradually builds up to this peak for most of the song, and then rather quickly fades out at the end. One of the things I used to do early on in my writing, was sketch out a quick graph to help guide me as I arranged all the parts. I'm not graphing the parts of the song, but rather how the energy or excitement of the track progresses over the length of time. For instance:
You can see that the song starts out rather tame, builds up for the first few minutes, almost peaks, goes into a drop section, then comes out of the drop into the peak, after which we slowly bring back down the energy level. Hopefully all in a way that's interesting enough to make people want to listen to it all over again!
Using a chart like this as a guide can really help you visualize how the song will look when laid out in your DAW. And yes this is music and we should be using our ears as well as our eyes, but let's be honest and just admit that we're going to visualizing a lot of this work initially. And remember this is just one way to lay out a song, I'm just working through one mock-up example to show you the process. There's millions of ways of progressing in a song when you think in terms of energy, so don't take it that this form is the best or right way to do it.
The next step is to translate that energy chart into the parts of our song we've already written. As mentioned earlier, we already know that the peak of the song will be when all the parts are playing, so that section of the song is already done! Let's look at the beginning of the song next, because we already know that this is where very little is going to be happening in relation to the peak. This means that we won't have too many audio parts playing at the same time, so pick a couple parts of your loop elements that are interesting, but not the core parts of the song. Maybe a percussion part, some pads or ambient sounds, something to catch people's attention but not give away the plot so to speak.
I find that 16-32 bars is usually a pretty good starting place for the intro, though some dance music will probably work better with more bars. Another consideration for club music is that you want to to have a really strong rhythmic element in the intro too, so that the DJ can cue up the track and beatmatch it if necessary. In this case it's not uncommon to have the kick drum playing right from the get go, though in my example I'm going to wait a bit to introduce it. Here's what I came up with for this song:
As you can see, a few element of the drums, and some of the less important (or less 'strong' might be a better term) synth elements. The same concept applies for the ending of the song too, we only want a few elements playing at once, something to ease people out of the tune and bring the energy down a bit. And again, for club music you'll likely want to leave rhythmic elements playing until the end, to give the DJ something to mix out of. Here's where we are now with the example:
In this example I've also gone ahead and create some locators to help guide me on where the various parts of the song will be, just some visual reference points. Again, refer to your energy chart to see where these should go. I tend to stick with putting song sections in multiples of 8 bars (i.e., 16, 24, 32, 48, etc) as it's what most people are comfortable with when listening, it's where they expect changes to happen. No reason you can't do something different if you feel the need though!
Sticking with the low energy parts of the song for now (and because it's easier to shape these parts initially), let's look at the drop section next. Again, the point here is to bring the energy of the song way down, so that when we come out of the drop and into the peak of the song, it's super exciting and really gets people's attention. So remove all but a few audio parts for the drop for now, play around with different combinations of parts, using the mute buttons in your DAW to help you audition which parts work better together than others. You want to give people a taste of the peak to come, but still tease them with the bare minimum.
You can see in my example below that even though I have only a few elements playing together during the intro, ending, and drop sections, I take care that I use different combinations of sounds for each section. This helps to keep things from being too familiar and repetitive sounding in the lower energy portions of the song.
Already starting to look like a real song I think!
Let's tackle the build down to the ending of the song next, as it's a relatively short section, and will leave us with only the build up to focus on later. Again, I'm taking away parts that relieve some of the energy of the song, without dropping it so much that it feels like the end of the song yet. A lot of times this is where I'll take away the main lead sound, or use a different (and less strong) version of it. Also be aware of sounds you've maybe had playing for too long already, you don't want people to get sick of hearing the same sound throughout the entire song.
As always, listen to what you're doing, and don't be afraid of trying out different combinations of sounds to see what works best. Don't just listen to the section on it's own either, listen to how it transitions from the previous part of the song, and how it transitions into the next portion as well. Sometimes you need to spend a lot of time doing this, trying to make the song flow from one part to the next without sounding like it's too abrupt of a change. Take your time, play around to get the best results. Here's where we stand now:
All that's left now is to work on the build up leading into the drop, usually the hardest part IMO. The point here is to really hook people into your song, but again without giving away all the elements yet. I'd leave the main lead parts until the peak, focusing on adding more parts in slowly, while making sure that no one part gets too repetitious. This is usually where I start listening to the song from front to back, making sure that I'm not sounding too much like the peak or build down sections during the build up.
Keep in mind that when you introduce stronger elements into the song, they will likely over-shadow weaker parts. Or worse, the weaker parts will distract from the stronger ones giving them less impact or making things sound cluttered. For example, I introduce a strong part called "Sync Lead" at bar 49, and at the same time I take away the guitar, some percussion, and some synth sounds to make this part stand out more. It also helps to keep the energy a bit lower during this part of the song, to avoid getting too 'peak' sounding yet. I then add back a few of these parts right before the drop, to build it up a little before we go into the drop. Here's the completed Build Up:
So there we go, our song is all done, right?
Well.... no actually. Though certainly a lot of people do stop at this phase and wonder why more people don't like their song. You see, this is just the rough outline of the song, it's basic shape and not really a finished product in my mind. The song sections are very roughly defined, and the transitions from each part to the next are abrupt and frankly probably a little boring. The real key to getting a song to sound interesting and making people want to listen to again and again is how you work the transitions between sections, how you build up to them and make the whole song flow as one piece of music.
This is where you focus on adding fills, doing effects and synth automation, anything you can think of to keep parts of the song not only from sounding too static, but also working them so that they lead up to the next section of the song. You want people to anticipate what's coming next, even if you plan on throwing them a curveball and doing something they DON'T expect. This is what makes a song interesting, when you can get people to subconsciously predict what will happen they become immersed in the music, they feel like they are the ones shaping it to some extent.
There's so many countless ways to do this, I can't even begin to touch on them, so I'll just give some pointers:
- Use drum variations and fills. In slower parts of the song, don't be afraid to use different variations on your main drum sounds. Less notes in the pattern is one way to do this, as is using quieter drum hits. Use drum fills to signal a change in the song, a quick snare roll in the last measure before you switch sections for instance.
- Record new effects sounds. Use some synth swells, fading in the volume over a measure or two to sort of lead people up into a new section. Record a drum hit with a lot of reverb, then reverse it so the sound of the reverb swells up to the transition. Don't just lead into transition points either, make sure you lead out of them as well. Let that drum hit fade away using the same reverb, it makes things more natural sounding.
- Automate something like filter cutoff. It's a cliched thing to do, but it works well when you slowly open up the filter cutoff to add interest as the song progresses. You hear it all the time coming out of drops and into the peak part of a tune, it's a simple way to go from dull to exciting, literally. Try other parameters too, and even try doing it in real-time via the knobs on your synth or controller, you don't need to draw all these changes in via automation. In general I'd say I do this one the most, to try and keep any one sound from getting too repetitive sounding and keep it constantly evolving, even if only a little.
- Spoken word samples. Added at just the right place and not repeated ad nauseam, they can be a great tool to signal a transition (i.e. "here we go").
- Effect changes. Try using different effects on the same part in different sections of the song. Maybe a little bit of flanger in the build up, but some heavy chorus in the peak. Be creative, but don't go overboard, subtlety is the key.
- Volume automation. Use the volume of each part to your advantage. Even slight changes in the volume can help to signal a change coming up. Or instead of just stopping or starting a sound, fade it's volume in or out to make the transition less abrupt.
I think you get the idea. It's these little bits of ear candy that can help everything gel, and add interest not only over the course of the song, but moment to moment. There's some songs where I try and add something every 8 bars, so that the whole song flows and people hear something new each time they listen. More than anything, you want to keep the listener INTERESTED! If they get bored, you're number 1 priority has not been met, no matter what type of music it is.
So listen to your song many times all the way through, try and spot parts where things are too similar sounding for too long. Chances are if it bores you even a little, or you find your attention wandering, it'll happen even quicker to other people. Focus on those areas, try different sound combinations, or add something there to make it more exciting. Listen to how the song flows as a whole, does it meet the requirements you outlined at the very beginning? If not, is it better this way, or do you need to rework it some more?
Take your time, and take breaks often to get back some objectivity. Stop every once in awhile for a few hours and come back to it with fresh ears. Always listen to the entire song after these breaks, don't just jump right back in to working on a small section. Make sure the tune as a whole is working, before you focus on any one part. Remember, that is the goal, to make a song that's interesting, not just a few bars that are interesting. Otherwise, we're right back where we started :)
On a more personal note, if this guide (or any of my other guides) has helped you in your music making, please consider a small $1 donation via pay pal to the email address below. Even a dollar here and there really helps me and my family out more than you can realize. Thanks, and I hope you find this guide useful.
Peace and beats,
I'm the owner of Inner Portal Studio, a Seattle-based facility with over 10 years experience providing quality mastering and mixdowns of electronic music for producers around the world. I have been writing, releasing, performing and DJing electronic music of all genres for over 20 years, and currently I'm one of the few independent artists with music for sale on Addictech.com. You can find my blog, original music and DJ mixes, as well numerous audio production related tutorials available for free at http://tarekith.com.
Feel free to pass this document on as you see fit, though I ask that you do not modify it from it's current form, and give proper credit. If you see any errors, please let me know so I can correct them asap.