Why Is This Here?
There's a question you can ask about any object that instantly reveals whether it was designed at all.
There's a question you can ask about any object that instantly reveals whether it was designed at all: why is this here?
Pick anything. A button on a screen. A handle on a door. A field on a form. If the person who made it can answer immediately — and the answer traces back to someone the thing is for — it was designed. If the answer is some version of "it was already there," "the framework does that by default," or "someone asked for it once," it wasn't designed. It accreted.
This is the real distinction in design, and it's not the one people usually talk about. We tend to think of design as a quality of appearance — whether something looks considered, polished, tasteful. But appearance is downstream. The actual variable is how many decisions were genuinely made. A beautiful interface where nobody decided anything is just a well-dressed accident. An ugly tool where every tradeoff was weighed is designed, in the way that matters.
Which means the core skill in design is noticing. Taste — the ability to look at two options and know which is better — only operates on decisions you know you're making. Most bad design isn't the result of bad choices. It's the result of choices nobody realized they were making. Taste tells you which option is better; noticing tells you there was an option.
Noticing is hard for a structural reason: decisions hide. Every deliberate choice costs attention, argument, and the risk of being wrong. Defaults cost nothing. The framework ships with a layout. The last version had this feature. The competitor does it this way. Each of these is a decision you don't have to make — which feels like efficiency, and sometimes is. But those unmade decisions compound. A product is thousands of small choices, and if you only consciously make a hundred of them, the other nine hundred get made by whoever you copied, or by nobody.
Nobody is the worst designer of all. Decisions made by nobody have no purpose to serve, so they serve none. And this is visible from the outside. Some software feels like a mind is behind it — every screen seems to know why you're there. Other software feels like a committee's sediment. The classic demonstration is the Google homepage. Its competitors were portals: news, weather, horoscopes, stock tickers, each panel the residue of some past deal or initiative. Google was a logo, a box, and two buttons. The irony is that the emptiness began as an accident — the founders barely wrote HTML, and early users sat staring at the page, waiting for the rest of it to load. What turned the accident into design was what happened next: Google defended that emptiness against every team that wanted a tile on it. The origin was an accident. Keeping it was a decision, renewed for twenty years. Meanwhile the portals kept accreting, and users could feel the difference without being able to name it. Intentionality leaks. So does its absence.
Here's the uncomfortable part: not deciding is also a decision. Ship the framework's default and you have decided it's right for your users — you've just decided without thinking. There's no neutral ground. The choice isn't between deciding and not deciding; it's between deciding consciously and deciding by accident. Much of what we call "technical debt" or "UX debt" is really decision debt: choices made by accident that now have to be re-made on purpose. And re-making a decision costs far more than making it, since first you have to discover it exists.
Once you start looking, you find accidental decisions everywhere, made by everything except a person. The form has twelve fields because the database has twelve columns — the schema decided. The error message is cryptic because the engineer wrote it while debugging — convenience decided. The onboarding explains the product's structure rather than the user's problem — the team's mental model decided. In every case a real decision got made, with real consequences for real people, and no human was present for it.
The largest of these absent decision-makers is the org chart. Conway's law is usually quoted as a curiosity: organizations ship their communication structure. Read through this lens, it's a warning. If you don't decide what your product's structure should be, your company's structure will decide for you. The seams between teams become seams in the interface. The product becomes a map of the company instead of a map of the user's problem. And like all accidental decisions, this one is invisible from the inside — everyone involved made locally reasonable choices. Accretion never looks like accretion while it's happening. It looks like progress.
What forces noticing, oddly, is constraint. A team with infinite time and scope makes fewer real decisions, not more, since nothing requires them to. A constraint — one screen, one week, one sentence of copy — makes every element fight for its place, and fighting for a place is exactly what a decision is. The first iPhone keyboard is the canonical case: with no physical keys, there was nothing to inherit. Every pixel of it had to be argued into existence, and it shows. The constraint didn't limit the design. It guaranteed there would be one.
Subtraction works the same way, for the same reason: nothing defaults to removal. Things default to staying. Every feature that survives a redesign survived either because someone fought for it or because nobody looked at it. A team that never removes anything isn't being careful. It's not noticing.
There's an obvious objection here: you can't decide everything. A product has too many parts; you'd never ship. True. But the answer isn't to spread your attention thinly across the whole thing. It's to be deliberate about where you're being deliberate. The best designers I've worked with don't make more decisions than everyone else — they make their decisions in the places users actually feel, and consciously accept defaults everywhere else. Accepting a default on purpose is a decision. Accepting one you never noticed is not. The difference is invisible in the artifact and enormous over time: the person who chose a default knows when to revisit it; the person who inherited it doesn't know it's there.
If there's one practice that follows from all this, it's embarrassingly simple: ask the question about things that already exist, not just things you're adding. New things get scrutiny automatically — they have to justify themselves to get in. Old things have squatter's rights. A periodic audit where existing elements must re-justify their existence does more for a product than most redesigns, because a redesign usually changes how the inherited decisions look rather than re-making them. The audit isn't really a process. It's noticing, scheduled.
And the lens doesn't stop at products. A career has defaults too. So does a week. The job you're in, the meetings on your calendar, the city you live in — some of these you chose, and some of them happened, and from the outside the two are indistinguishable. Why is this here? If you can answer, it's yours. If you can't, it belongs to whatever decided it for you. Most people, like most products, are mostly inherited.