What We Lose When We Optimize Everything

I spend most of my working life optimizing things. Shaving milliseconds off a response time, cutting a database query from four joins to two, redesigning a pipeline so it does in one pass what it used to do in three. It is satisfying work, and I mean that sincerely — there is a real, physical pleasure in watching a system get leaner and faster under your hands. But somewhere in the last few years I noticed I had started applying that same instinct to things that were never meant to be optimized, and I want to write honestly about what that cost me before I noticed it happening.

It started small. I began tracking my mornings the way I track application performance — sleep score, time to first coffee, minutes to "deep work." I told myself this was discipline. In some ways it was. But discipline slowly curdled into something else: a low, constant hum of self-surveillance, where every hour of the day was implicitly being graded against some theoretical maximum-efficiency version of itself. A walk was no longer a walk. It was a missed opportunity for a podcast that would make me "1% better." A conversation with a friend was a context switch with a cost I was quietly resentful of, even as I smiled through it.

The thing about optimization as a mindset is that it is very good at making local improvements and very bad at asking whether the metric itself is the right one. In software, we call this Goodman's law territory, though most of us just call it "gaming the metric." You optimize support ticket resolution time, and suddenly agents are closing tickets before the customer's actual problem is solved, because the clock is what's being measured, not the outcome. I had, without quite meaning to, turned my own life into a system being gamed by a version of myself trying to satisfy a metric — busyness, output, hours logged — instead of the actual thing the metric was supposed to be a proxy for, which was something closer to a life well spent.

What snapped me out of it, a little, was a much older engineer I used to work with, who had spent almost two decades maintaining systems that predate almost every framework I've ever used. I asked him once how he decided what was worth optimizing and what wasn't, and he said something I still think about: "You optimize the things that are on the critical path to a goal you actually care about. Everything else, you just let be slow." He wasn't talking about life. He was talking about a legacy billing system. But it landed somewhere else in me entirely.

Because the truth is, most of what makes a life feel rich is not on any critical path. The unhurried dinner. The conversation that wanders nowhere useful. The hour spent reading something with no professional application whatsoever. None of it optimizes anything. All of it is, by the metrics I had quietly adopted, waste. And yet I would not trade a single one of those hours for the marginal productivity gain I told myself I was chasing.

I don't think the answer is to reject optimization wholesale — that would be its own kind of overcorrection, and frankly I still love the craft of it, still get a genuine thrill from a system running clean. The answer, I think, is closer to what my old colleague said: decide, honestly and in advance, what is actually on the critical path of the life you want, and guard everything else from the optimizer's instinct entirely. Let the slow things stay slow. Let the inefficient conversation run long. Not everything that matters needs to justify its own existence in terms of output.

I still track fewer things than I used to. Not because tracking is bad, but because I got tired of living inside a dashboard of my own making, one where I was simultaneously the product manager, the engineer, and the increasingly exhausted user. Technology gave me the instinct to measure everything. It's also, if I let it, capable of teaching me when to stop.

If there's a version of this worth passing on to anyone building things for a living, it's this: the discipline that makes you good at your job is not automatically the discipline that makes your life good. They can look identical from the outside — the same relentlessness, the same refusal to accept "good enough." But one of them is in service of a system. The other one is supposed to be in service of you. It's worth checking, every so often, which one you're actually running.