Training Your Eye for Motion

Software got cheap. Building things is easier than ever. AI writes code, Figma has templates for everything, and anyone can ship a product in a weekend. The technical barrier that once separated amateurs from professionals has almost disappeared.

So what's left?

Judgment. The ability to look at two versions of something and know which one is better. And then knowing why.

This is taste. And it's becoming the most valuable skill you can have.

Press both buttons below. Same duration. Same scale value. One uses spring easing with overshoot, the other uses linear timing. Completely different experience.

The difference isn't subjective. One feels alive, the other feels mechanical. Developing the ability to see this difference and know how to fix it is what this post is about.

#The Myth of Personal Preference

When people say "taste is subjective," they usually mean they don't want to argue about it. It's a convenient escape hatch. If taste is just personal preference, then there's no right or wrong. Everyone likes what they like.

But that's not how it works.

When you improve at any craft and look back at your old work, you don't think "this is different." You think "this is worse." And you can explain why. The timing was off. The easing was wrong. The element moved when it should have stayed still.

Show someone two landing pages. One from Vercel, one from a random SaaS template. They'll pick the Vercel one as "better" even if they can't explain why. Show a developer two code editors. They'll have a strong opinion about which one feels right. These aren't random preferences. They reflect something real about quality that most people can sense, even if they can't articulate it.

Taste isn't arbitrary. It follows patterns. And patterns can be learned.

#Why This Matters Now

Think about headphones. When the first portable music players came out, having any headphones at all was the feature. Sound quality barely mattered. Just being able to listen to music on the go was enough.

Now every company makes headphones. They all play music. So what makes someone choose one over another? The weight distribution. The texture of the ear cups. The satisfying click of the folding mechanism. The way the case opens. Details that used to be invisible became the reason people pay three times more for one product over another.

Software is going through the same shift. For a long time, just shipping a working product was enough. Features mattered. Scale mattered. Now everyone has features. Everyone can scale. AI makes it even easier.

What makes a product stand out is how it feels. The micro-interactions. The loading states. The subtle animations that tell you something happened. The personality embedded in every transition.

Raycast doesn't win because it has more features than Spotlight. It wins because using it feels different. Figma didn't take over design because Sketch couldn't do the same things. It won because every interaction felt more considered.

Companies are realizing that good taste creates good products. And good products are becoming rare because most people aren't developing their taste. They're just shipping.

#Taste Can Be Trained

Here's the thing most people miss. Taste isn't a gift. It's a skill. Like any skill, it improves with deliberate practice.

The pattern is simple. Expose yourself to great work. Study it closely. Ask why it works. Then try to recreate it.

When you see an animation you like, don't just admire it. Slow it down. Record it. Watch it frame by frame. Ask yourself:

• What's the duration?
• What's the easing?
• Which properties are animating?
• What happens before? After?
• Why did they choose this movement over others?

Then build it yourself. Not copy-paste. Actually build it from scratch. You'll discover things the original creator thought about that you never noticed.

This is how musicians learn. They don't just listen to great music. They transcribe it. Note by note. They internalize the choices that made it great.

#Finding Your Sources

You need a curated list of work to study. Not random things from the internet. Work from people who have proven they understand motion.

Some places to start:

Products that feel great

Raycast. Things 3. Craft. Cron (now Notion Calendar). Amie. Bear. Use these products slowly. Notice the animations. They've spent enormous effort on details most people never see.

Individual practitioners

Find designers and engineers who share their process. People who post breakdowns of interactions they admire. People who explain why certain animations work.

Rauno Freiberg's work is worth studying. So is the team at Basement Studio. The Vercel design team. The folks behind Sonner and Vaul. These people think deeply about motion.

Open source libraries with opinion

Radix. Shadcn. Sonner. Vaul. These aren't just component libraries. They encode decisions about timing, easing, and interaction patterns that you can learn from by reading the source.

#Seeing What Others Miss

When your taste improves, you start noticing things everywhere. The slight delay before a tooltip appears. The way a progress bar slows down as it approaches completion. The overshoot on a toggle switch.

This can be overwhelming at first. Suddenly every interface feels either delightful or painful. You can't unsee the details.

But this is the point. You're training pattern recognition. Your brain is building a library of "this works" and "this doesn't." Over time, you won't have to consciously analyze everything. You'll just know.

The same thing happens to musicians who develop an ear. They hear a chord and immediately know it's a minor seventh with a flat five. It's not magic. It's exposure and practice compressed into intuition.

#The Difference Between Rules and Judgment

This series covers a lot of rules. Ease-out for entrances. Ease-in for exits. 200ms as a baseline duration. Spring physics for interactive elements.

These rules are useful starting points. They'll get you most of the way. But taste is what tells you when to break them.

Sometimes ease-in for an entrance is correct. Sometimes 400ms is better than 200ms. Sometimes a linear animation creates exactly the mood you need. Rules tell you what usually works. Taste tells you what works here.

The best designers I know have internalized the rules so deeply they don't think about them anymore. They just feel when something is wrong and know how to fix it. That's not ignoring the rules. It's having run the rules so many times that they've become instinct.

#What AI Can't Replace

Large language models can write code. They can implement animations. They can even follow best practices. What they can't do is tell you if the result feels right.

"Make this button animate on press" is easy. Any AI can do it. "Make this button feel satisfying to press" requires judgment. It requires someone to look at the output and say "no, that's not quite it" until it is.

This judgment is taste. And it's the thing that will matter most as code becomes cheaper to produce.

The people who develop their eye will be the ones who guide the machines. They'll be the ones who know when the output is good enough and when it needs another pass. Technical skill will be table stakes. Taste will be the differentiator.

#Building Your Library

Start collecting examples of motion you admire. Screenshots don't work because motion is time-based. Use screen recordings. Build a folder of clips you can reference.

When you add something, write a note about why you added it. "I like this" isn't useful six months later. "The way the notification slides in and bounces slightly at the end" is useful forever.

Over time, this library becomes your personal reference. When you're stuck on how to animate something, you'll have dozens of examples to draw from. When you're trying to explain to someone else why something should feel a certain way, you'll have concrete examples to point to.

#Conclusion

Taste isn't magic. It's accumulated judgment from years of paying attention to what works and what doesn't. It comes from studying great work, breaking it apart, and rebuilding it yourself.

The people who win in a world of commoditized code are the ones who can look at something and know if it's good. Not because of opinion. Because they've developed the pattern recognition to see what others miss.

Start today. Find one animation you love. Describe why. Build it. Then find another.

Your taste is waiting to be trained.