Config
Decide and move on
I’ve fallen into traps working on a single feature on my website, or on a micro detail like kerning or a button state. Something impactful, but small. Some of these lasted months.
What made me stop is that I realized I was losing steam. I was wasting time. And how did I know it was a trap? Because I kept swapping between ideas. I’d have one, then swap back to the other, swap in and swap out. I’d comment things out. That’s how I knew it was indecisiveness, not care.
Something I’d never walk away from is whether it feels right. That goes for the quality of my components, and even my writing, where I’ve fallen into traps of making something tighter, or more brief, or more human. If it doesn’t feel right, I stay with it. That’s what makes it a trap, because staying with something until it feels right, or until it passes the brief, is part of what design is.
If you ditch it too early and say it’s done, you leave things unexplored. It’s like carving marble, a Michelangelo thing. You’re trying to set the angel free, so you have to keep going. It might seem like grunt work until you find it, and then you have to say to yourself: if I had stopped yesterday, if I’d stopped five minutes ago, what would the outcome have been?
A set time
In client work on a retainer, I can’t fiddle with something for hours. So I do it at the source, on the things everything else is built from, get it completely right, and never come back to it. A design system, a template, the writing rules on my website. Measure twice, cut once. Get it right the first time, so that it ships perfect.
Versus working with a template you haven’t spent the time on. Every time you execute it, you see the same issue in your face, and you inevitably have to fix it again. You didn’t put the time in, so you put it in later.
So for me, setting a time comes down to a few questions. How important is the thing? How much time do I actually have? What else is coming up?
The other side
The opposite anti-pattern is ignoring the micro details, and focusing only on the macro, on what sells, on what people see.
Most people don’t see the micro details. But for those with a discerning eye, and not just the designer or the professional or the writer, those things stand out. In a sea of good enough, they’re what sets a thing apart. The outputs of Claude, of AI agents, of a template that’s widely used: they’ll look great. Perhaps that’s good enough. Dig deeper and it’s homogeneous, the same as everything else. There’s no life. The micro details are the care you put into something, and in those outputs they’re not considered.
I don’t know if it bugs anybody else, but when I see something super thin drawn next to something with thicker lines, in a cartoon or a children’s book, it throws me off. It’s a balance thing for me, so maybe I’m just picky. Part of it is depth, but not all of it. Artists often scale things down, and I’ve also seen an anime where something in the background loses detail as it fades away but keeps the same line width. I find myself complaining in both cases. But when you draw things far away, you have to decide whether to lose detail, or to scale things down and lose consistency.
The same face at one stroke weight. Shrunk as it is, it keeps the detail and loses the balance. Redrawn small, it keeps the line and loses the detail.
With typography, even if you change a value and change it back to the default, at least you touched it. I think that’s the point. Everything should be touched at least once, even if it’s in the template and never again in the execution. We let go of the details no one will notice, but we don’t ignore them. We address them and move on.
That’s what happened with my own site. I was trying to get a single font weight across the site, which was easier said than done. It ended up being about visual weight, not a literal font weight. The big type looked way too thin next to my 2-pixel rules, so after a lot of tweaking I went back to weights that felt the same, rather than ones that were the same. I couldn’t tell you why. It just felt right.
The trap Staying with it until it feels right. Sep 2026 · 5 min read Subscribe
One shared weight
The small text goes thinner than the line.
The trap Staying with it until it feels right. Sep 2026 · 5 min read Subscribe
Weight follows size
The smaller the text, the heavier it gets, to match the line.
The same card at one weight, then with the weight following the size, so the small text and the button match the line.
Feeling your way to it is the art in design, and you don’t always have to say why. In Taste is triage I said clients need the why, and with client work they do. But that’s a balance too. Sometimes it’s fine to just like it.
So the anti-pattern shows up as design as commodity, as good-enough UX, as the same thing the other guy did. You don’t have to do something different. You just have to do it better.
When to stop
And the trap, again, is that sometimes better doesn’t matter. The art is placing your attention on what matters, using your taste to triage your decisions. This matters to me; will it matter to them? Does it feel good enough?
In most cases you can come back to it. In some cases you can’t, and that’s where you set a time, because it’s better to ship something good enough than not ship at all because you spent the time trying to make it, quote unquote, complete.
That’s the trick to knowing when to stop. That’s where taste comes in, and discipline.
Sometimes fiddling with something is the job. Those are gifts, moments where you can slow down and really zone in. But with so much pulling at our attention, and generation becoming cheap, that’s when we really need to step back. In some cases, direct, not produce. Keep our eyes on the bigger picture. Those moments of letting go are the ones that move us forward.