When is a page transition worth it vs just delaying the content?
On this page
A page transition is worth it only when it does work the content cannot do for itself, when it preserves context, shows a spatial relationship, or covers a load the user would otherwise sit through staring at a blank screen. Outside those cases it is pure delay, motion you have inserted between the user and content that was ready to show. The test is simple to state and easy to skip: does the transition earn the time it costs, or is it a curtain you are pulling in front of a finished room. If the content is sitting there ready, the honest move is to show it, not to make the user watch it slide in.
The reasoning is that a transition spends the user’s time, and time is the one budget you cannot refund. Every animated entrance, every slide and fade between screens, adds a few hundred milliseconds before the person can read or act. That cost is justified when the motion returns something in exchange. A transition that keeps an element in view as it moves from one screen to its detail view tells the user the two are connected, so they do not lose their place. A transition that fills the gap while data is genuinely loading turns dead waiting into something that looks alive. But when the destination is already rendered and waiting, the motion returns nothing. It just makes the user pay the toll on an empty road. The habit worth naming is “add page transitions to make navigation feel smooth,” which treats motion as a finish applied to everything rather than a tool used where it works.
A designer recognizes both sides in a mobile app. Tapping a photo thumbnail that expands smoothly into the full image, the thumbnail growing into place so the eye follows it without a cut, is a transition doing real work: it preserves the spatial relationship and tells you exactly which photo you opened. Compare that to a content site where every link triggers a half-second fade-to-white before the next page fades in, even though that page was prefetched and ready. The first transition orients you. The second one just makes you wait for text you could already be reading, and after the fifth navigation it stops feeling smooth and starts feeling slow.
This flips when “does work” includes covering a real wait, not only showing spatial connection. If a screen genuinely needs a beat to assemble, a transition that bridges that beat is earning its delay rather than inventing one, because the alternative is a flash of blank or a jarring pop-in. The line falls where the content is ready and the transition still insists on running. A short, restrained motion on a heavy load is defensible; a flourish in front of an instant render is not. Worth-it is about whether the user would have waited anyway, not about whether the motion looks nice in isolation.
When you reach for a page transition, run the earns-the-delay test before you build it. Ask whether the motion preserves context, reveals a spatial relationship, or covers a load the user would otherwise wait through blankly, and if it does none of those, cut it and let the content appear. Use a transition where it carries the user across a gap, and let ready content arrive without a curtain in front of it.