Finishing tracks: Why I Spent a Year on a Remix and Six Hours on a Hit

The remix took me the better part of a year. The track everyone actually remembers took an afternoon.

This is not me bragging, it’s one of the more useful things I’ve learned about finishing tracks in my twenty years of doing this. The piece I agonised over, picked apart, and rebuilt over and over again isn’t the one that connected. The one I let go on is. And once you’ve seen this mechanism in play a few times, you start to suspect your inner critic is lying to you about what matters.

The remix that took a year

I did a remix for Silicon Sound’s “Nexus 6” many years ago remix of which nearly broke me. I loved the original. Loved it. And that was exactly the problem. Every time I sat down with my version I wasn’t really asking “is this good,” I was asking “is this allowed to exist next to something I rate that highly.” Those are completely different questions, and only one of them lets you finish anything.

So I fiddled. I’d swap a lead, second-guess the arrangement, decide the low end wasn’t sitting right, come back a week later and undo all of it. Months of that. The track wasn’t getting better in any way a listener would notice, it was just getting moved around by someone who was scared to call it done. You can listen to where it eventually landed here on YouTube. It’s a good remix. It did not need eleven months.

The track that took six hours

“Talisman” was the opposite experience in every way. I sat down, the idea came, and I just followed it. No committee in my head, no comparing it to anyone else’s production, no endless A/B against a reference until the meaning drained out of it. Six hours, start to finish, and I let it go.

That’s the one that sat near the top of the Beatport trance chart for the best part of a month. The original mix is up on YouTube if you want to hear it. To this day it’s one of the biggest things I’ve made, and it took less time than I’ve spent choosing a kick on tracks nobody will ever hear.

I’m not pretending six hours is some magic number, or that speed is the point. The point is what was absent. The critic never showed up. And the track was better for it.

The inner critic is measuring the wrong thing

This took me years to say out loud. When you’re stuck at 90 percent, going round in circles, your inner critic is almost never measuring whether the track is good. It’s measuring whether the track is correct. Whether the production stacks up against the last thing you heard from a producer you admire. Whether the bass is EQ’d the way a tutorial told you it should be.

Those are real considerations. They are also not what makes a person reach for the volume knob. The average listener is not buying your sidechain settings or the surgical dip you put at 250Hz. They’re buying a melody that makes them feel something. The emotion is the product. The production is the delivery van.

I think a lot of us, me very much included, get this backwards because the part of the brain doing the judging isn’t really a music critic at all. It’s a fear response. I’m late-diagnosed AuDHD, and one thing that clicked into place when I finally understood myself is how much of my life has been organised around not standing out, not getting something wrong, not being the one who did the strange thing. Which is a genuinely funny way to be wired when your actual job is to make something that stands out. Creative work is the one arena where the whole point is to be the strange thing in the room, and there I was, terrified of exactly that.

You don’t need a diagnosis to recognise the mechanism. Most perfectionism in the studio is fear wearing a technical costume. It dresses up as “the mix isn’t ready” because that sounds responsible. Underneath, it’s “I’m scared this is different and different might be wrong.”

“But production is a real gatekeeper”

I can hear the objection, because it’s a fair one. Sloppy production does get passed over. Labels reject it, curators skip it, and standards are real. I’m not the guy to tell you production doesn’t matter, I’ve spent years honing mine and I take it seriously. There’s a barrier to entry in electronic music, and clearing it takes time, graft, and some talent. I’m not waving any of that away.

But notice what the barrier actually does. By the time you can produce competently at all, you’ve already cleared the technical bar most people never get over. Production becomes table stakes. Musicality, the melody, the feeling, the thing that makes someone replay a track at 2am, that’s the rarer commodity. So when you’re forced to choose between a track that’s musically incredible and one that’s immaculately mixed but says nothing, you take the musical one every single time.

And something I wish I’d understood sooner: there’s emotion in rawness. A production left slightly unfinished can carry feeling that a perfectly polished one sands right off. The roughness can be the point. So before you decide a track isn’t good enough on production grounds, ask whether the “flaw” is actually doing some of the emotional work.

How I tell the difference now

This is the whole game, separating “genuinely not ready” from “just scared.” I don’t always get it right, but here’s the question I ask myself when I hit that wall.

Is this lacking musically, or am I just comparing my production to someone else’s and flinching because mine sounds different?

If the honest answer is that the melody, the energy, the feeling are there, then the track is done and the rest is me stalling. Ship it. If the answer is that the core idea genuinely isn’t moving anyone, including me, then no amount of EQ will save it and I should go back to the writing, not the mixing. Either way, the move is to stop polishing. Polishing is what the critic does when it doesn’t want you to find out whether the song works.

The track you’re scared to release because it’s “not as good” as someone else’s production is often the one with something nobody else has: yours. Stop measuring it against the wrong ruler. Finish it, and let people feel it.

So next time you’re at 90 percent and the loop starts, try it. Ask the question, give yourself an honest answer, and then actually do the brave thing, which is usually just clicking export.