Everyone has superpowers now. If someone doesn't like how you've built something, they'll build it themselves. The hard engineering problem — the thing that used to be the moat — is commoditized. What remains is taste. Good taste plus good engineering instincts is the edge.
The design lifecycle
Design happens in phases, just like software development.
- Inspiration — You see something that moves you.
- Ideation — You start forming your own version of it.
- Iteration — You refine, test, and refine again.
- Result — Something ships.
Most people skip straight to iteration. They open a blank canvas and start pushing pixels with no foundation underneath. That's how you end up with things that work but nobody remembers. Taste is built upstream — in what you consume and how you think — long before anything gets committed to code.
Taste is a muscle
The good news is that developing taste isn't some innate gift. It's a muscle, and it's one of the easier ones to build. It just requires active practice.
You can do something as simple as scrolling Pinterest and studying UI components. Joining design discussions on Twitter. Watching how designers build products. Or just paying closer attention when you use products — not as a user, but as a student. Why did that feel smooth? Why did that annoy you?
Good design knows when to add friction and when to remove it. A loading state is a great example — it's friction, technically, but it keeps the user oriented. They know something is happening. They stay in the process. On the other hand, cutting unnecessary steps — removing a confirmation screen nobody reads, collapsing two pages into one — that's taste too. Knowing what to keep and what to kill is the whole game.
The practice doesn't have to be formal. It just has to be consistent.
Go where taste makers go
A chef samples many spices and tries to create new dishes. A builder tries many applications. They use things they'd never build themselves. They notice the moments that feel good — a transition, a layout, a micro-interaction that just works — and they isolate those moments. They save them. They share them to close the feedback loop and learn whether what they liked is actually good, or just personal preference.
The tools for this are Cosmos, Pinterest, Mobbin, Twitter. These are taste libraries. Similarity search matters here — algorithms feed you things on the fringe of your existing interests. Similar enough to resonate, different enough to stretch.
Most good designers have one or two people they look up to. Or one design style they're chasing — a north star they measure everything against.
Learn the names of things
Naming something is important. But describing something is even more important, especially when you don't know the name yet.
"That's a bento grid." "That's a staggered reveal." "That transition uses spring physics." When you can articulate what you're seeing, you can search for it, reference it, recreate it. Vocabulary is a design tool. The more words you have for what you see, the more precisely you can direct your work — and your agents.
Seasons
Designers have seasons the same way fashion does. One season you're deep into 3D and isometric work. The next you're drawn to ASCII, or brutalism, or integrating real-world photography into interfaces.
This isn't indecision. It's range. Good designers are always honing their taste — cycling through influences, absorbing new styles, letting them inform what they make. The breadth of what you've tried becomes the depth of what you can create.
The ones you can't forget
You recognize great design because you keep remembering it. Days later, weeks later — it's still there. Your brain was stimulated in a way that tickled something creative, and you can't let it go.
That's the signal. When a design haunts you, chase it. Study it. Break it apart. Figure out why it stuck — was it the color, the typography, the spacing, the motion? That feeling is usually the best indicator of a style worth emulating.
Complex, then simple
Designs start complex and become simple. First make it work, then make it elegant. The first version is messy — too many elements, too many ideas competing. The discipline is in what you remove. Simplicity isn't the starting point. It's the destination.
The bottleneck has moved
Before, the competitive advantage was solving the hardest engineering problem. Now it's understanding the user better. Engineers have built godlike coding abilities out of their own intellect, and those abilities are now available to everyone. The question isn't "can we build it?" anymore. It's "does it feel right?"
That puts PMs and designers in a strong position. And if you're a builder who wants to be a designer, now is the time — you can borrow taste through curation. Screenshots, saved references, design library components, video — all of it can be surfaced as context. You can approximate taste while you're still developing your own.
Go where the taste makers go. Observe the patterns. Learn the names of things. Emulate what came before you. Test a lot. Isolate features. Build a store of things you love — a Pinterest board, a Cosmos collection, a folder on your desktop. Show people what you're making and ask what they think.
Pay attention to the designs you can't forget. Then chase that feeling.