Most design work optimizes the interface and ignores the conditions under which it appears. That's backwards. By the time a user reaches your button, the environment has already done most of the work.

This is what gets skipped in most design work. The button, the label, the layout: all of it gets refined across dozens of iterations. The moment the user arrives at them rarely does.

In the late 1970s, a researcher named Roger Ulrich studied patients recovering from gallbladder surgery at a Pennsylvania hospital. Half of them had rooms overlooking a brick wall. The other half looked out at a stand of trees. The tree-view patients left the hospital nearly a full day sooner. They needed less pain medication. Their nurses filed fewer negative notes.

Ulrich published this in Science in 1984. Hospitals cited it for years in renovation documents. And still, most patient portal interfaces look like they were designed in a windowless room by someone who had never been a patient.

The interface lives inside a context. That context is doing work: shaping what users expect, what they fear, how much patience they have, what they're willing to read. Most usability work treats the context as fixed and the interface as the variable. We test the button and neglect what the user went through to get there.

A hospital registration form at a scheduled 9 a.m. visit and the same form at 2 a.m. in an emergency department are not the same. The person the form was designed for, calm and attentive with a few minutes to spare, doesn't exist in that room. The environment hasn't made that persona available. What's there instead: fluorescent lights, a muffled PA system, a crying child, a triage wristband. The form lands on top of all of that.

Casinos understood this long before anyone ran a usability study: you don't change a person's behavior at the machine, you change the room around it. No windows, no clocks. Carpet patterns designed to push the eye off the floor and back up toward the machines. But the slot machine isn't the interface, and neither, really, is the building. The whole city is. Las Vegas is engineered street by street to hold you inside a single state of mind, and the machine is just the place where that state finally cashes out.

You could redesign every slot machine on the strip and not move the numbers meaningfully. The environment has already done the work.

And the loop doesn't end at the city limits. The same logic runs on the device in your pocket, and it runs both ways: the environments we build shape what we reach for, and what we reach for shapes the next environment we build. We are constructing the spaces we live in, and they are constructing us back. It is being done on purpose, not by a hidden hand but by everyone who ships a screen, writes a flow, or lays out a room. The question isn't whether it's deliberate, but whether it's conscious, and what it's for.

This is what goes undesigned in most products. The team owns the interface. Nobody owns the conditions under which it appears. And those conditions shape more than how people feel: they determine what people can conceive of doing. In a high-friction context, certain options simply aren't available, not because the user lacks the skill or the willingness, but because the environment hasn't made that mode of attention possible. You can't reach for something that the moment won't let you reach.

So the work starts earlier than the button: at the moment of arrival, not the moment of completion. Ask what state the user is in when the product first appears: not which persona they belong to, but what they've probably come from. What they're probably carrying. What they can still see clearly. What kind of attention they have left.

But preparing for the moment is only half of it. Whatever you ship doesn't just sit inside someone's environment; it becomes part of it, the conditions the next person arrives into. So the real question is what it's for. The same craft that engineers a casino to extract can engineer a room that leaves space for attention, for patience, for a decision someone actually means. We don't only account for the context; we get to build it, and to build it better than we found it. Make the room better and it makes us better in return. That is the loop, turned a notch in the right direction.

A button matters less than what brought the user to that moment. The context is something we are choosing, every day, whether or not we admit we're the ones designing it.