Ed Rex is Founder and CEO of Jukedeck as well as being a composer. He originally wrote this for his Medium blog.
When I entered the lecture hall, I was a musician.
By that I mean I’d sung in choirs, played in bands, and pored over musical scores for 23 years of my life. But try going to an Introduction to Computer Science lecture and not emerging with a burning desire to learn to code. It can’t be done. So when I left the lecture hall, I started a startup. It was October 2010. And I’ve been doing it ever since.
The change has been eye-opening. I’ve been able to compare, first hand, the way the music and technology worlds operate. And they couldn’t be more different. There are the obvious differences, like the music industry’s very public history of moving slowly and resting on its laurels; its reluctance to adapt to change nearly killed it at the dawn of the iTunes era. But we’ll take the fact that it needs to embrace new technology as a given. What I’d suggest is that it needs to adopt whole new ways of working — new approaches to the way music is actually created — that will mean musicians can produce better music, faster.
And the way to do this is to start acting more like startups.1. It’s about the listener
The Lean Startup
movement has assumed an almost Biblical authority in software circles. Its guiding principle can be summed up in four words: It’s about the user.
What does that mean? It means that if no one’s using your product, it may as well not exist — so everything hinges on making sure you understand what your users want. This is why MailChimp has an Evernote notebook for user research in which every single employee documents every conversation they ever have about Mailchimp. With anyone.
There can be a tendency to think the opposite is true in music: the artist’s job is to find their true expression, and if the listener doesn’t get it, the fault is with the listener. But this just isn’t how music works, and it never has been. In the past, music was usually written for a specific patron, congregation or audience, and if composers didn’t write music those people liked, they’d find themselves out of work. The same applies, albeit in a different guise, in 2014 — the songs that top the charts are written by people who know what the public wants to hear, and are really, really good at producing it.
Shaping your music according to what people will like isn’t selling out — it’s how to write good music. After all, there’s no objective measure of what ‘good’ music is, other than what’s popular. If you accept this — that it’s about the listener — you can start going about actually figuring out what it is the listener wants. And that’s where the Lean Startup principles come in.2. Create Minimum Viable Tracks
It used to be the case that software companies would spend a year or more working towards a big release, moving through various phases of development one at a time and launching into the unknown with a ‘no going back’ approach. This is known as the Waterfall method. It’s now been widely replaced by Agile, which involves building fast, releasing a Minimum Viable Product
as early as possible and learning from feedback.Music is stuck in a Waterfall world.
A composer might spend years on a piece, working in complete solitude until the last note is in place. What if, at the premiere, no one likes it? Shouldn’t that have been found out sooner?
Enter the Minimum Viable Track. When you’re recording an album, why not play the first takes to as many people as you can, before you embark on weeks of editing and post-production? If no one’s going to buy it, doing that post-production is a waste of time — time you should be spending writing new tracks. And you need to find that out as early as possible.3. Iterate & Pivot
Startups are obsessed with iterating: constantly trying new things and experimenting with new features until one proves popular.
There’s no reason, as musicians, we shouldn’t do the same — that is, once we’ve got the early versions of our music in front of our audience, start responding to their feedback. If there are bits they like, concentrate on those — if there are bits they don’t, scrap them and try something else in their place. I tried this recently with a song I’m writing — instead of spending months working on it on my own, I put together a rough and ready demo on my iPad, sent it to a couple of people, and immediately found out there were a couple of lyrics they thought let the rest of it down. So out went those lines, and in came a series of new ideas, which I kept changing until they had the desired effect — the approval of these early listeners.
To improve is to change; to be perfect is to change often." Winston Churchill
What if you find from your early demo that people basically don’t like what you’re doing, full stop? While this is never nice to hear, it’s better to find out early than keep going with something no one will like. In this case, in the tech world, you’d pivot
— that is, change course and try something entirely new. Pivoting isn’t a sign of failure — it’s a badge of honour, a sign that you’re willing to take tough decisions to get to a product that people actually want. Indeed, some of the biggest tech companies out there performed early pivots. Twitter? Originally a place to subscribe to podcasts
. Starbucks? Started out selling espresso makers
So if your early feedback tells you you’re writing music people don’t like, why not try something else? There’s nothing wrong with pivoting.4. Use A/B Testing
There’s a huge movement in technology towards A/B testing, which means putting out different versions of your product and seeing which drives more user engagement. It’s becoming so instrumental to the way technology is made and improved that there are entire companies
built on providing A/B testing as a service.
In music, there’s no A/B testing. Why not? There’s nothing stopping you putting out two early versions of a track — with a couple of different choruses, for example — and seeing which gets more likes and shares. You can then go with the version that people like more. Using A/B testing means you can employ the wisdom of crowds instead of making each and every decision on your own.5. Don’t stop at launch
There’s a prevalent feeling that, once written and recorded, a musical work is set in stone — to change it is to take a hammer and chisel to something sacred. But this is actually a relatively new idea. The composers of old chopped and changed their pieces well after the first performance — Handel, for example, continually revised his operas, adding new arias and rejigging roles to accommodate the various singers who happened to be available for each performance. And jazz tunes, with their emphasis on improvisation, are reinvented every time they’re played.
The technology world works like musicians used to. Launching a product doesn’t mean it’s complete — updates and entirely new versions will continue to be released on a regular basis. Look at the difference between the Facebook of 2005:
And the Facebook of today:
Musicians could gain a lot from readopting the mindset of our predecessors and being more willing to make changes post-release: our music would stay interesting, our style would stay relevant, and our fans would stay engaged.6. Get efficient
If you want to see efficiency in action, spend a week in a startup’s offices in San Francisco. The rate at which these guys churn out products and features is mind-blowing. The first version of Twitter, for instance, was built in just two weeks
Compare that to the music world (not all of it, but a decent amount of what I’ve seen). I’ve been in recordings where people stroll in two hours late, argue over mixing levels for half the day, and take a cigarette break every thirty minutes.The recording process needs to be more efficient than this.
The greatest musicians of the past operated more like startups do today. Bach wrote over 1,100 pieces of music
during his lifetime. Handel composed the Messiah
in 24 days, started Samson
a week later, and completed the latter in a month. These guys didn’t waste time. Today’s musicians — at least, a not insignificant subset of us — could massively increase our chances of success by upping our efficiency.7. Work in sprints
One of the most useful tools in Agile software development is the sprint — a fixed period of time in which certain goals are set to be achieved. Generally, a team will start a new sprint every 2 weeks, and review their progress at the end of that period. This means the huge task of creating a product from scratch is tackled in manageable chunks, and you have a regular check-up as to how well or badly things are going.
I’ve recently been trying this with the music I’m writing — instead of sitting down at the piano whenever the mood takes me and seeing what I come up with, I’ve been setting myself specific goals to achieve in a set amount of time. Finish the chorus by the end of the weekend, that sort of thing. And I’m finding it massively useful: I got that chorus done, and am now ready to record it. (That will be the next sprint.)
Ultimately, imposing deadlines on yourself, in the form of sprints, breaks the endless cycle of improving and perfecting that makes musicians everywhere reluctant to release their work to the world. (I’ve written an entire post on how deadlines are the one true antidote to perfectionism
.) Sprints rein in your perfectionism and make sure you actually get things done.
It’s tempting to think that the creative process can’t, or shouldn’t, be subjected to the principles of business. But startups are hugely creative themselves, as proved by the kinds of things they come up with every day. So if these tools are good enough for them — and were good enough for the musicians of the past — isn’t it at least plausible that they could be useful for us, too?
As musicians, maybe we’d do well to be more lean.