Why motion fails more than it helps
Motion is the most expensive design decision you can make, and most teams make it casually. An engineer spends two days building a staggered card entrance animation. A designer spends a week refining easing curves on a sidebar. A PM ships a loading skeleton with a shimmer effect that serves no functional purpose. None of these decisions go through the same scrutiny as a headline change or a CTA color test.
The reason motion fails isn't because designers don't care. It's because the default question is 'does this look good?' instead of 'what does this do?'
"Motion that doesn't communicate something isn't design. It's decoration. And decoration has a cost: attention, performance, and trust."
The three questions
I use a simple filter for every animation I consider. It has to answer at least one of three questions truthfully:
Does it show where something came from? Entrance animations that reveal content from its source location — a drawer sliding from the edge, a modal expanding from its trigger, a tooltip appearing beneath the cursor — orient the user spatially. They make interfaces feel physical.
Does it show where something is going? Exit animations matter as much as entrances. A card that flies off the stack in the direction of deletion confirms the action. A nav item that collapses toward the destination sets expectation.
Does it confirm something happened? State changes — a button press, a form submit, a toggle — often happen faster than a user can register. A subtle spring on confirmation, a brief color flash, a checkmark that draws itself: these are functional, not decorative.
If an animation can't answer one of those three questions, I cut it. Not reduce it. Cut it.
The Cash App Friday problem
When I designed the Cash App Friday campaign microsite, the brief included the word 'energy' fourteen times. Energy was everywhere — in the moodboard, in the copy brief, in stakeholder feedback. And the temptation with an energetic brief is to express it through constant motion: things bouncing, scrolling fast, animating on entry and exit.
The problem with that approach is that everything competing for attention cancels itself out. When everything moves, nothing moves the user.
What we shipped instead was a page where most things were static, and two or three moments had genuine kinetic energy — moments tied to audio cues, moments that rewarded a specific user action. The restraint made those moments land harder.
The brief was about energy. The solution was about focus.
"The most energetic design I've shipped was also one of the most restrained. Restraint is what gave the energy somewhere to go."
A practical test
Before shipping any animation, run this test with your team. Remove it. Show both versions to someone who hasn't seen the design before. Ask them to describe the interface after 30 seconds. If the static version produces descriptions that are meaningfully less accurate or confident, the animation is doing functional work and belongs. If the descriptions are roughly equivalent, it's decoration.
Most animations don't survive this test. That's not a failure — it's a feature. Every animation you cut is performance you keep, engineering time you reclaim, and cognitive load you remove from the user.
Ship fewer animations. Make the ones you ship matter.