Read this short list once, then look away and try to recall it: table, chair, lamp, 7, shelf, desk, rug. If you’re like almost everyone, the thing that survived the wipe is the number. Not because a 7 is more interesting than a lamp, but because it was the one item that didn’t belong to the pattern the rest of the list agreed on. Your memory spent almost nothing on the sixth piece of furniture and a disproportionate amount on the single thing that broke rank.
This is one of the oldest and sturdiest findings in memory research. A psychologist named Hedwig von Restorff described it in the 1930s, and it has been called the isolation effect ever since: in a field of similar things, the odd one out gets encoded harder and recalled better. It shows up again, in a faster form, in vision. When you scan a screen, a single element that differs from its neighbors on one clean dimension — one red dot in a field of green ones, one filled shape among outlines — seems to pop out at you before you’ve consciously searched. You don’t scan item by item looking for it. Difference does the finding for you.
Now hold that next to the screen you’re building.
A first-time user lands somewhere in your product and, for a fraction of a second, their attention is up for grabs. There is exactly one thing you most want them to do next — start the trial, create the first thing, hit the one button that begins the story. And here is the trap almost every careful team walks into: they love their product, so they make everything nice. The secondary link is a crisp button. The tertiary option is a crisp button. The help text is bold. The banner pulses. Every element has been lovingly turned up to a nine.
The problem is that emphasis is not an absolute quantity. It’s a contrast. A bright element is only bright relative to duller ones around it. When you raise everything to a nine, you haven’t made everything important — you’ve made nothing distinct, and von Restorff’s effect has nothing to grab. The eye lands nowhere in particular, the user does a slow conscious scan instead of an instant pop-out, and the one action you cared about is now just another thing in a crowd of equals.
So the build move is not the one people reach for. The instinct, when the primary action isn’t getting clicked, is to add to it: make it bigger, brighter, throw an animation on it. That sometimes helps, but it’s the expensive path, and it escalates — because the moment your neighbor makes their button glow, yours has to glow harder. The cheaper, more durable move runs the other way: subtract emphasis from everything that isn’t the one action. Make the secondary choice a plain text link instead of a button. Let the help sit quiet and grey. Give the tertiary stuff no weight at all. You are not dimming your product; you are building the field of sameness that lets the single important thing pop.
There’s a phrase designers use — “if everything is bold, nothing is bold” — that usually gets nodded at and ignored. Von Restorff is the reason it’s true and not just a slogan. Salience is a budget, and it’s zero-sum. Every element you promote quietly demotes the one you actually care about. The most confident screens I’ve seen are mostly quiet: a lot of calm, near-invisible scaffolding, and one thing that unmistakably wants to be pressed.
Ask of any screen you’re about to ship: if a stranger glanced at this for half a second and then it vanished, which single element would they remember? If the answer isn’t the one you’d have chosen, you don’t have a copy problem or a color problem. You have too many things trying to be the 7.
Liked this? Get the next one in Working Theory.
Going weekly in August (it's in beta now). One genuinely interesting read on building, the brain, and the science most people missed.