I’ve written, recorded and released into the wild about a dozen completely original songs, but I still don’t really have a process for writing them, or at least not a conscious one. This post is an attempt to figure out what works for me and what doesn’t. I’m not, by basically any conventional measure, a “successful” songwriter, nor do I pretend it’s anything other than a part time hobby for me. So don’t construe this as advice or think I imagine myself to be a model for other songwriters to emulate. I’m just thinking through what, if any, process I personally have, to try to have a little more control over it in the future.
I have a lot of half-formed lyrics sitting around in notebooks and a lot of snippets of music sitting around on my hard drive, or, more rarely, on sheet music. Some of these I like well enough, but very few of them ever end up becoming songs I like enough to finish, much less release.
On the other end of the spectrum, I’ve had a few song ideas just come to me—this is a total cop-out phrase, but bear with me—and then sort of unfold into entire songs very quickly and easily. A couple notable recent examples of this: “Where No One Has Gone Before,” whose intro verse melody sprang up more or less fully formed during a long bike rideI distinctly remember stopping at a park to record the melody into my phone. I’m self-conscious about recording in the best of conditions; doing so in public means I had what seemed like a guaranteed Eventual Actual Song. And I did!
, and “The Ice World,” where the first verse’s lyric just fell out of my emotional state at the time with melody attached. In both of these cases the rest of the lyric followed fairly quickly, and in a way that felt inevitable or at least very natural. Any other melodies in the song were pretty straightforward to derive from that first flash of inspiration as well.
“Inspiration” is, of course, the biggest cop-out word of them all. What the hell is “inspiration?” It seems un-analyzable: we imagine it as a bolt of lightning, powerful, random, impossible to command. I’m not going to try to lasso that lightning here, at least not yet. But some flashes of inspiration lead me to fully fleshed out songs, and others sputter out. What’s the difference?
The lyrical seeds that grow most easily into entire songs tend to combine an emotion or feeling with a metaphor or analogy robust enough to create a sort of feedback loop: the metaphor describes how I feel somehow, and other associated thoughts and emotions reinforce that metaphor by leading to a new perspective on it or introducing a second one that I can tie in. Then the additional detail or clarity on the metaphorical side leads to a deeper or more nuanced understanding of my real-life emotional state, and so on… and before long I’ve got a whole lyric written.
That’s kind of the large-scale conceptual framework that seems to be reliably fruitful for me. Harder to pin down is what makes an initial seed hooky enough to warrant expanding it into a whole song.
The only reliable technique I’ve discovered for evaluating “hookiness” is to refrain from recording or writing down a melody until I’ve figured out how to expand on it. If I forget it before then, it wasn’t a good enough hook. This is risky, but at the same time it seems to work well. I’m not necessarily trying to write earworms, but “enough to stick for a few days” seems like a good lower bound on catchiness. It also leaves the melody a little space to lead to another section of the song, before I pin it down and trap it under a chord progression. That tends to constrain the directions I can take it, which is essential after a certain point but not something I want to do prematurely.
Expanding fragments into actual songs needs to feel fairly intuitive to me. I don’t like to feel like I’m fighting with my songs, and it always seems like the results get worse the more I have to do so. If I have a verse and a chorus and can’t figure out how to tie the two together, it’s often a sign that the weakerHow do I determine what’s “weaker?” I kind of just have to trust my own judgment from being a music fan for about 20 years. Some red flags: lyrics that are too vague or use too many words to say too little, or that are metrically awkward; melodies that feel clichéd or obvious; too-contrived or (sometimes) too-basic chord progressions; things I don’t enjoy listening to myself; things I dread having to record in such a way that they sound good.
one needs to go or be radically re-written so it fits better. Having a strong overarching concept helps here, as does having some kind of rhythmic or harmonic framework. Any of those things help constrain where I can go from what I have, which is desirable after I have enough seed material to sustain a whole song. At that point, anything that enables me to throw out bad options more easily makes the whole process less of a struggle. Though I don’t have much technical proficiency in any instrument, which immediately limits my options to “things I can actually play,” modern DAWs still provide a glut of sonic options that need to be pared back to avoid analysis paralysis.