I sorted music for a living. Sometimes ten seconds was enough

Then I'd sit down to my own work and lose fifteen minutes choosing a track.

For about three years I worked for a stock music library, and most of the job wasn't what people assume.

I found the artists myself, solo performers and small bands, then explained how the library and licensing worked. The ones who said yes sent me their catalogues. I uploaded the files and started listening.

That's where the real work was. Track by track, and for each one the same questions: what genre, what mood, what instruments, and the one that mattered most: what kind of video would someone put this behind. Cooking. Wedding. Travel. Corporate. That's what the tags were for. Someone editing a video searches that way, and a track tagged wrong may as well not be in the library at all.

Whether I personally liked it never came into it.

Sometimes ten seconds was enough. A few bars in and I already knew where it went. Other tracks I'd have to sit with, replay, work out whether the shift halfway through ruined it for the thing it otherwise suited. Occasionally something was genuinely good and I'd quietly add it to my own playlist, but that was rare and beside the point. Mostly it was work. Music wasn't something I was collecting.

I don't do that job anymore. What I do now has nothing to do with music, which is part of why it took me so long to notice what happened next.

I'd sit down to work and put something on, and fifteen minutes later I'd still be skipping tracks. Not listening to them, just sampling, a few seconds each, looking for the one that would be right. Some days I'd look up and realise I'd been doing that for a while and hadn't opened the thing I was supposed to be working on.

Which is absurd on its face. I used to place a track in ten seconds. Then I couldn't pick one to work to in fifteen minutes.

The flattering explanation is that the job left me fussy. Trained ear, high standards, can't settle for background noise. I believed that for a long time.

It doesn't survive contact with what the job actually was. At the library I was answering a question that has an answer: where does this go? Wedding or travel, guitar or synth, decide, next one. Alone at my desk I was asking something with no answer at all. What's the perfect track for right now? A question with no answer can be asked indefinitely. That's the appeal. Every track was one more small decision, and every decision was another chance not to begin.

It's the same move as "I'll start now, or in ten minutes, or after coffee." It just doesn't look like avoidance, because it looks like preparation.

Gloria Mark and her colleagues at UC Irvine spent years studying how information workers switch between activities. In their field observations, self-initiated interruptions were roughly as common as external ones. My case wasn't quite the same, since technically I hadn't started anything to interrupt, but the distinction stuck. Distraction doesn't always arrive from outside. I was handing myself permission to reconsider, over and over, and calling it taste.

What helped wasn't better music. It was taking the choice away.

I use the same track to start every work session now. One I have no opinion about. Not a favourite, not something I'd defend to anyone, just something that doesn't ask me for anything. Press play, start. Nothing to evaluate, so nowhere to go looking.

I don't want to oversell it. If the task itself is miserable, no ritual fixes that; I still put things off, just less often at the starting line. Plenty of people work fine with a new playlist every morning, or in silence. What helped was having a default, one fewer decision between sitting down and starting.

I'd be curious whether this has a version in other people's work. Not music, but the thing you do that looks like setup and isn't. Reorganising the config before starting. Deciding which ticket to open first, then deciding again. Renaming the branch.

Mostly it left me wondering how many of my other preferences are like this one. Not taste. Just a decision I've dressed up well enough that I don't have to look at what it's for.

I currently work on Nedio. The product deals with a related problem, reducing the friction between sitting down and starting.