“That doesn’t look good.” Four words that are hammered into every design career like nails in a coffin. Usually in meetings where the tension is so thick you could cut it with a drawing board. No why. No explanation. Just that one sentence, sitting there like a deity, expecting you to instantly know what it means.
* Spoiler: *
You don't.
What always annoyed me most about it: What looks good to me might be completely different from what looks good to others. Does that mean one of us is wrong?
This thought kept me up at night for a long time. When I was just starting my career, one of my lecturers gave me an answer I've never forgotten since: Design has to work, he said seriously, as if this was a concept that was just being successfully explained to me.
* Spoiler: *
You don't just know.
I wish I could say that was an exception. It's not. Taste is treated in this industry like an innate talent – something you either have or you don't. Either you're born with it, or you're always in the wrong meeting. What nonsense.
It's a skill. A trainable ability.
And like all other skills, you get better at it by putting in the time, not by waiting for it to just happen to you one day.
The Ira Glass moment that every designer knows (and hates)
Ira Glass – Journalist and podcast god behind This American Life – once said something that has since been quoted in every other design Reddit. Mostly by the very people he was referring to, who haven't quite grasped it yet.
In essence: When you start designing, you already have taste.
You know what looks good. You immediately see when something isn't working.
The problem is: You can't do it yet. The gap between "This is crap" and "I can do better" is the hellish cycle most of us spend about three years in.
What Glass describes there is aesthetic judgment. The ability to evaluate a design before it's finished. Sounds innate? It's not. It only sounds that way because people who have it are reluctant to admit they spent hundreds of hours scrolling through design sites (aka Pinterest, Dribbble, Behance, Mobbin).
Here's the plot twist: The people around me whom I considered magically "tasteful" had no fewer doubts than I did. They just had more mileage. Seen more designs, analyzed more crap, made more mistakes.
"I don't know what's wrong, but I know when it's right"
Heaven forbid that sentence. Almost everyone has said it. Me too.
"I don't know exactly why, but it just doesn't work." And yes, as an observation, that's not wrong at all. Sometimes your gut says "nope" before your brain knows why. The problem is: If you stop there, you come across as someone who appreciated a work of art without understanding it.
The real difference between "I have a good feeling about it" and "I know what I'm doing" is one thing: articulation.
Can I explain why this isn't working? Is it the font weight? The whitespace-to-area ratio? The color temperature that just doesn't fit?
If I can't say that, I'm not the person in the meeting making smart decisions. At best, I'm just a first filter.
Paula Scher likes to say in interviews that design happens "in seconds." That's true. But those seconds are the result of about 30 years of work.
Every time she says that, junior designers imagine it as a magical moment. But: That's not talent. That's practice. The second is fast because 30 years stand behind it.
How to actually train your taste
(or: The Inconvenient Truth)
Step one: Exposure – and intentionally so.
Not scrolling Pinterest on the subway while simultaneously staring at your phone. But: Take your time. Look closely. Understand why something works or doesn't. Stefan Sagmeister has spent years analyzing beauty in design. His approach is extremely unsexy: He looks for patterns. He asks what the things he likes have in common. He tries to distill that. That's diligent work with a notepad, not "artistic genius."
Step two: Critique.
Read design critiques – not because you agree with everything written there, but because it teaches you the language to even discuss designs. Good designers don't write about their work as self-aggrandizement. They write like an accountability report. What was it supposed to be? What did it become? Why? If you read this regularly, after a while you won't just form an opinion on designs. You'll gain the vocabulary to defend that opinion.
Step three: Friction.
Taste is trained faster when it's challenged. When someone says "I find this more beautiful" and you have to explain why not – without resorting to flimsy personal preference arguments, but with real reasons. That's uncomfortable. It's also very effective.
What this has to do with Art Direction (and why it hurts)
Art Direction without a well-defined taste leads to a very specific kind of meeting. You know this meeting. The client sits there, and instead of bringing a clear position, you show five variations. Because you yourself don't know which one is best. The client doesn't get happier in that moment. They become insecure. They had hoped someone would bring a stance, and instead they get a selection problem where they have to make the decision themselves. That's not service. That means you're the wrong person in the room.
Good taste in Art Direction doesn't mean you're always right. It means you have a position – and can explain it. "This variation, because the type hierarchy is clearer, the contrast works there, the color choice matches the visual language." The client can reject that. But they know what they're dealing with.
Diana Vreeland – legendary Vogue editor, who saw herself as an authority – is said to have stated that elegance is not a state of being, but a decision. I'm not sure if she said it exactly like that (the quote is one of those that always float around in design blogs), but the idea is true. Taste is also a decision. Daily. And like all decisions, you get better at it when you make them consciously – instead of hoping it will just "happen" to you one day.
Why this is more important now than ever before (and almost funny, if it weren't so sad)
AI tools now deliver results that aren't wrong.
Layouts that are clean.
Typography that doesn't immediately make you cringe.
Color palettes that don't alarm anyone.
That's impressive. I also work with it. But here's the cliffhanger: A tool that produces functionally usable results doesn't replace the ability to know if the result is the right is.
What I've learned working with AI tools – and this is something that surprised me: The hard part isn't generating. The hard part is refining. Looking at the output, seeing that something isn't quite right, and then translating that into a language the tool understands. "Make it less corporate" isn't a prompt. That's the language of someone who has no idea what's bothering them. "The font hierarchy is too homogeneous, the contrast between headline and body text is missing" – that's a prompt. That's also: a clear indication that you need to know what's going wrong.
The difference between "I somehow know this isn't right" and "I can say why this isn't right" – that's not a matter of talent.
That's aesthetic judgment. And yes: that's precisely why art direction doesn't automatically disappear as AI tools get better. On the contrary. Someone needs to be there who knows what they want. And most importantly: why. Someone who can write good prompts. And ultimately: can judge if the result is correct. That's not a workflow problem. That's taste.
What remains
Back to my teacher and his "You just know that."
I don't think that came from convenience. I believe he looked at it himself for so long until it became invisible to him – just like you eventually stop hearing your heartbeat. The result of a 15-year process looks like an innate ability if you say it fast enough.
That's the problem with "innate taste": It describes the end stage and pretends that's the beginning. If you eventually believe that judgment is simply there or not, you stop working on it. Self-fulfilling prophecy. In both directions.
The truth is less magical and much more practical: There's no difference between someone who "just knows" what works and someone who can explain it – except for the fact that the latter has looked closely enough to develop a vocabulary for it. Talent here is a nice word for work.
And if you shouldn't find that optimistic: That means you can start. Now. Today. Without waiting for your taste-gene to be activated. That means the person in the meeting next week who says, "That doesn't look good" – you could be the person who says, "Yes, and here's why."
TL;DR
Taste develops through exposure, not predisposition.
The difference between intuition and judgment is the ability to provide reasons.
Showing five variations because you have no preference yourself is not customer service. It's insecurity.
Reading design criticism develops a language for evaluations – that's underestimated.
Whoever cannot explain why something isn't working is only filtering. Not deciding.
.png)
