Some Stuff I’ve Learned Writing Music for Advertising—Why I Keep Doing This

What really makes or breaks a relationship with a freelance composer working in the advertising business is the revision process. Composers who can make changes in a quick and friendly way rise to the top of the list, whereas those who constantly present resistance and debate fall to the bottom of the list.

Written By

Ravi Krishnaswami

Revision & Expansion

In my previous post, I looked at the very thorny issue of how we communicate with clients to understand what they are looking for and why. Communication is probably just as important as composing chops when it comes to successful collaborations with clients, but I often tell my students that what really makes or breaks a relationship with a freelance composer is the revision process. Composers who can make changes in a quick and friendly way rise to the top of the list, whereas those who constantly present resistance and debate fall to the bottom of the list.

I certainly believe in taking a stand at times. There’s a basic level of respect that artists deserve, and clients who don’t listen to their vendor’s ideas are only limiting the creativity of their own work in the end.  If I feel strongly about an approach, I will also advocate to at least include it as an option alongside the client’s preferred approach. But similarly, composers who assume they always know best and can’t possibly improve their work by incorporating client feedback are closing themselves off to expansion and growth.

Composers who assume they always know best are closing themselves off to expansion and growth.

While in the heat of the moment I’m often loathe to admit it, hindsight makes it abundantly clear that there have been many counterintuitive feedback requests that have pushed my work to new places, opened up creative doors that I assumed were closed, and revealed to me things I didn’t know I could do. When the artistic voice in my head screams “that’s impossible!” after receiving a request, it’s the business voice that mutters “just try for their sake” that pushes me forward, often into a better place.

Take, for example, the music I wrote for a Cadillac Super Bowl commercial. The original demo included the key elements of the final music that the clients loved, such as the tremolo lithopone and lullaby synth melody.

Cadillac Chrome Couture Demo Mix from COPILOT on Vimeo.

But the original drum and percussion section was more tribal sounding with a quarter note pulse, pushing the whole piece slightly into the world of dance music. One of the big gut-check issues with the spot was whether it was skewing too far towards a female-only demographic given its fashion show milieu. So this eventually made its way into anxieties about my music, and I was tasked by the agency’s creative director with revising the drums until they got more muscular, primal, live, and raw sounding, with less of a groove. There was no time for studying these hunches. (There rarely is.) Revisions like this one just happen when creative people are motivated by a deadline and an open-ended problem.

These and other changes to carve sections out and create more surprising moments (perhaps they worried about the visuals not being impactful), and edits to follow changes in timings, which seemed like an arbitrary hassle to me at the time (my original demo was perfect, couldn’t they see that?), pushed me to write something that was ultimately weirder and more attention-grabbing. Boy did I appreciate that when it came on in the middle of the biggest TV slot of the year, with no voice over or additional sound design cluttering up the final music.

Here’s the final version:

Cadillac “Chrome Couture” from COPILOT on Vimeo.

Don’t get me wrong, I’ve spent countless hours executing bad ideas. Many come from clients that know just enough about writing music to cause trouble. They need to hear the idea not working before they give up on it. Many others are simply compromises between two strong creative positions with different motivations and stakeholders. “Splitting the difference” or “finding the midpoint” usually means watering down in the end. Nonetheless, I’ve grown to really embrace feedback as a moment where I get pushed and challenged, and I’m ultimately grateful for those moments as they make me learn new tricks. You can’t get too possessive about your music in this industry. You have to be completely 100% emotionally invested in what you are doing when you are doing it, because that is the level of artistry that’s required to be successful. But once the music leaves your computer, it takes on its own life, and you must simply wish it luck and offer support when needed.

As a young composer, I got down about having to try ridiculous things mentioned by folks with no idea what the process would entail. Now I see it as a challenge: a great composer can take any note on a piece and address it so well that the client feels like a genius for suggesting it. Many clients, after all, want to feel like they are adding something creative to a project, that it wouldn’t be the same without their ideas. If you show your client that you care about their ideas and won’t leave them hanging, you have a pretty good shot at another project.

Sometimes as a composer you are creating notes for musicians to play, sometimes you are creating a space for collaborators to play in.

There’s another very practical advantage to bringing the client’s ideas directly into the piece and making them work. In situations that involve a lot of stake holders and layers of bureaucracy, and when agency teams have listened to many other options for a spot, you can turn that person into an advocate for your composition because of the sense of ownership that comes with contributing ideas. Sometimes as a composer you are creating notes for musicians to play, sometimes you are creating a space for collaborators to play in.

Why I Keep Doing This

If you are reading NewMusicBox, it is likely you’ve seen rumors about our new president eliminating the National Endowment for the Arts. We could be entering a dark period for federal arts funding.

I’m going to use a controversial economic term, but stay with me: music for advertising allows money from companies to “trickle down” into the arts community. These projects provide us an income and hone our skills while we continue to pursue our passion projects. The year I scored that Super Bowl spot, for a product I personally disliked (due to its poor fuel economy), was also the year I wrote and produced my band Charming’s third album, without the need for label support. Not only was I able to live comfortably in the midst of recording an album that was at best a niche product, but I had free studio time at the jingle house.

What I do could easily be seen as a selling out to corporate interests. By definition it is. Yet it doesn’t feel that way.

The greatest joy I get from my job as lead composer and creative director at COPILOT is being able to bring work to my talented friends. What I do could easily be seen as a selling out to corporate interests. By definition it is. Yet it doesn’t feel that way when I’m also hiring a diverse group of composers, musicians, singers, and engineers for projects.

Session work took a real hit when high-quality sample libraries became ubiquitous, production schedules shrank, and budgets imploded. Live recording became thought of as a time-consuming luxury. But now, when I meet a musician, the first thing I ask is, “Are you set up to record yourself?” With a good computer, quiet room, microphone, audio interface, studio headphones, and software, a musician or singer can be available for session work at a moment’s notice. While I love working face-to-face in the studio when I can, the ability to work remotely has opened many smaller projects up to live recording, mostly due to how much quicker things can get done.

What’s Ahead

In the ensuing debate following the election, I heard many pundits talking about how it’s not trade deals and immigration that will kill American jobs, it’s automation. And lo and behold, last week The New York Times published a piece about a company called Jukedeck. Apparently they’ve developed an artificial intelligence system for writing original music for media projects. There were rumors for years in the jingle industry about composers and programmers dabbling in this area, so it wasn’t a complete shock.

I’m not terrified yet. In some ways, an AI system for cranking out music doesn’t seem like a far leap from the crowd-sourcing scale of library music. Can an AI system keep up with current trends in composition and scoring? Can an AI system make the kind of mistakes or breaks from convention that create new trends? Can an AI system understand comedy? Can an AI system move beyond a single emotion or style and combine things in new, unexpected ways? The answer to all of these questions is, of course, “We’ll see.”

A composer’s job is not just to crank out music.

But if I’ve tried to communicate anything through these posts, it’s been that a composer’s job is not just to crank out music—it’s to understand a problem, understand the trajectory and context of a project, and to build relationships and trust through communication.

If AI figures out how to do all of that, its final hurdle would be authenticity. As a “jingle composer,” I know this challenge well. No matter how inspired my work is, if it’s coming from a company that specializes in jingles and a guy that does that for a living, it will seldom carry the weight of any artist’s work for a certain percentage of clients. Having written music in both contexts, I don’t believe that my spirit suddenly dies on commercial projects and soars on my passion projects, but I do believe that in the razor thin margins of subjective judgement about music, perception becomes reality. In fact, it became common for jingle houses to sign a few well-known artists to their roster to add luster, or—even more cynically—to invent identities for successful underscores to lend them more credibility. With the proliferation of sync licensing, I certainly see this bias going away down the road, but I have to imagine that when it comes to AI, most human beings would like to know that another human being wrote the music they are using. At least for now.


It’s been really fun trying to form coherent sentences around a half-life of instincts and lessons from the trenches. I hope you’ve enjoyed reading them here, and wherever your career is headed, good luck and be yourself. One last bit of advice: Attaching files to emails… please… just stop. The kind folks at WeTransfer, Dropbox and the like, would love to offer you some free services.