Why you love one app and hate another (even when they do the same thing)

I promised in the last post that I'd come back to talk about emotional design. Here I am, and I brought Don Norman with me.

Have you ever downloaded two apps that do exactly the same thing, same function, similar screens, and still only ever open one of them? The other one works just fine, no errors, no freezing. It just never won you over. And that doesn't show up in any product metric I've ever seen.

Every time I teach this, someone asks the same question: why does a beautiful app make me trust it before I've even really used it? Well, because we decide we like something long before we understand it. The emotional decision comes first. The rational justification comes later, running behind to explain what we already felt.

Beautiful design isn't fluff. Fact. Usability alone doesn't cut it. Fact. But there are still people treating "beautiful" and "functional" as rival teams fighting over the same brief, as if only one could win.

The person who ended that fight was Don Norman, back in "Emotional Design: Why We Love (or Hate) Everyday Things", from 2004. And he settled it in a way I still use today to explain digital products to my students. Let's go.

Quick rundown of the three levels at which we process design, according to Norman:

• Visceral: the first impression, before any conscious thought. Color, shape, texture, the glow of the screen. This level is almost animal, inherited from a brain that needed to decide fast whether something was safe or dangerous. Today it decides, in milliseconds, whether an app looks trustworthy or cheap, whether a screen looks cared for or abandoned. You judge before you know why, and then spend the rest of the session validating that judgment.
• Behavioral: the level of actual use. Does the button respond fast? Does the flow move or does it stall at every step? Does the task end where it should? Here aesthetics alone can't hold anything up anymore, the product has to deliver. It's the most studied level in UX, the one we run usability tests on, the one that shows up in task-completion metrics. But it's also the coldest of the three: it measures efficiency, not attachment.
• Reflective: the level of meaning. What does using this product say about you? What memory or identity does it carry? It's the slowest level to kick in, because it demands awareness, comparison, history. And it's also the longest-lasting: an ugly screen is quickly forgotten, a badly solved task is quickly forgotten, but the feeling of belonging to something stays.

The catch is that these three levels don't always walk together. A product can win on one level and lose badly on another, and that's exactly where the challenge lives for anyone designing interfaces.

Bringing it into our universe: visceral is that first second of screen, before the user understands anything about the navigation. Palette, icon, breathing room in the layout. Behavioral is the actual flow: clear feedback, zero friction to complete a task, nothing stalling along the way. And reflective is the reason you keep one specific icon on your phone's home screen, or feel proud enough to screenshot a screen and send it to the group chat.

There's a classic example Norman himself uses: the Juicy Salif lemon squeezer, by Philippe Starck. Gorgeous, became a design piece in museums, terrible at actually squeezing lemons. Visceral winning by a landslide over behavioral, and it still sells to this day. Now take the opposite: Craigslist. Ugly since forever, no palette, no icon, no breathing room in the layout whatsoever, and it's been standing for decades because the behavioral level runs like clockwork. Visceral zero, behavioral ten out of ten. Two products, two different levels carrying all the weight alone.

And the reflective level? That's the one that shows up most in digital products today, just in disguise. Spotify Wrapped hands you a data point about your year in music, but the real product there is something else: something worth posting, something that tells the world who you are through what you listened to. Duolingo plays a similar move with the streak: keeping the run of days alive became less about learning a language and more about not letting down the person you became after thirty days in a row. In both cases, the level carrying the product is the reflective one, from start to finish: identity, consistency, the story you tell about yourself.

The book's central thesis fits in one sentence I steal straight for my classes: attractive things work better. Norman opens that conversation with two studies done with ATMs, one in Japan and one in Israel: same functions, same buttons, and still the versions with the more attractive layout were perceived as easier to use. The explanation has less to do with ease and more to do with patience: we forgive a product we love much faster when it fails, and we come back even after a bad experience.

Want to apply this to your product? Audit the three layers separately, one at a time, no mixing.

First, the visceral: open your home screen on its own, with no context, as if it were the first time anyone has seen it. What's the reaction in under a second? Second, the behavioral: take your three most-used flows and walk through each one from first tap to the end, counting how many times you hesitated or backtracked. Third, the reflective, which is the hardest to measure and the one almost everyone skips: ask what the person feels at the exact moment they finish a task in your product. Do they leave proud, relieved, indifferent? If you can't answer that last question, that's where the problem lives, not in the button that "could be prettier".

A kiss, and see you next time!