Figma added 3D transforms, and the first wave of reaction was the usual demo dazzle: cards tilting on hover, elements floating in perspective, interfaces that suddenly have depth. It looks impressive in a thirty-second clip. The harder question, the one that matters for anyone actually shipping a product, is whether depth on the canvas is a genuinely useful design capability or a feature that mostly produces things that look cool in a portfolio and hurt in production.
The honest answer is both, and the split is sharp. 3D in interface design has a small number of genuinely good uses and a large number of tempting bad ones, and the difference between a designer who uses it well and one who overuses it is entirely about knowing which is which. This piece covers where 3D transforms actually earn their place, where they backfire, the performance and handoff costs nobody mentions in the demos, and a verdict you can act on.
What 3D transforms do
Figma added 3D transforms, and the first wave of reaction was the usual demo dazzle: cards tilting on hover, elements floating in perspective, interfaces that suddenly have depth.
Set the baseline plainly. 3D transforms let you rotate frames, vectors, and text on the z-axis with native controls and a live preview, giving elements depth, tilt, and perspective rather than the flat, two-dimensional plane interfaces have mostly lived on. The transforms stay fully editable and export to CSS, and they connect through MCP so spatial design can match shipped code. Combined with Figma Motion, they can animate, enabling things like card flips, perspective menus, and depth-based transitions. One caveat worth stating up front: at launch this is a preview feature available via waitlist rather than to everyone, so treat it as early rather than settled.
That's the capability. The question is not what it can do, the demos answer that, but when using it improves a product rather than decorating it.
Where 3D actually earns its place
There is a real, if narrow, set of cases where depth genuinely helps.
Tactile microinteractions. A subtle tilt on a card as a user hovers or presses can add a sense of physical response, a small tactility that makes an element feel touchable. Kept subtle, this reads as craft, the kind of quiet detail that separates premium-feeling UI from flat, generic interfaces. The word doing the work is subtle: a few degrees, not a dramatic flip.
Product and object showcase. When the product itself is visual, hardware, a physical object, a 3D asset, showing it with real depth and letting it rotate serves genuine understanding. The depth is about the thing being presented, which is a legitimate reason for it to exist.
Spatial storytelling in brand and marketing moments. On a marketing page, an onboarding sequence, or a brand moment, a sense of depth can create atmosphere and guide attention in a way flat design can't. These are low-stakes, attention-seeking surfaces where a little spectacle is the point, and depth can be part of the craft rather than a distraction.
Notice the pattern: the good uses are subtle, purposeful, and mostly live on brand or presentational surfaces, not in the core functional UI a user works in every day.
Where it backfires
The bad uses outnumber the good ones, and they're tempting precisely because they demo well.
Depth in functional UI. Adding perspective and tilt to the screens a user works in daily, dashboards, forms, tools, almost always hurts. It adds visual noise, competes with content for attention, and makes a working interface feel like a toy. Functional UI wants clarity and flatness; depth there is friction dressed as flair.
Parallax and motion overload. Layering 3D depth with scroll-driven parallax and constant motion produces the kind of page that impresses for five seconds and exhausts on the sixth. It's the design equivalent of a font with too many weights: capability used because it's there, not because it's needed.
Disorientation and discomfort. Aggressive 3D motion, spinning, dramatic perspective shifts, depth that moves as the user scrolls, can genuinely disorient people, and for users with vestibular sensitivities it can cause real physical discomfort, the same accessibility concern that governs all motion. This isn't a matter of taste; it's a matter of not making some of your users unwell, and it ties directly to the reduced-motion discipline that good motion design requires.
The costs nobody mentions in the demos
Two practical costs sit underneath the aesthetic question, and both matter for anything that ships.
Performance. 3D transforms and the effects around them are more expensive to render than flat design, and heavy use, especially animated depth, can make a page feel sluggish on lower-powered devices and drain battery on phones. A design that runs smoothly on the designer's laptop can stutter on a mid-range phone, which is where much of the real audience is. Depth is not free, and the cost lands on exactly the users least equipped to absorb it.
Handoff. A 3D effect that lives on the canvas still has to be built in code to ship, and depth and perspective effects are meaningfully harder to implement well than flat layouts. What looks like a quick tilt in the design file can be a real engineering task with its own performance tuning, so a designer casually adding 3D everywhere is quietly adding to the build cost, the same gap between what a design tool shows and what production requires that shows up across the board.
The verdict
Use 3D transforms the way you'd use a strong spice: sparingly, deliberately, and mostly not in the main course. On brand and marketing surfaces, in genuine product showcases, and as subtle tactile feedback on interaction, depth can be real craft that lifts the work. In the functional UI users live in, as layered parallax spectacle, or anywhere it risks disorienting or slowing users, it's a gimmick that costs more than it gives.
The tool being available is not a reason to use it. The best designers will treat 3D transforms as an occasional, purposeful accent and leave their working interfaces flat, clear, and fast, which is the same restraint that separates considered design from the everything-at-once look, a theme that runs through the Liquid Glass versus Material Expressive debate about depth and expression in general.
Start Monday
Before you add 3D to anything, ask one question: is this a surface where a user is trying to get something done, or a surface trying to make an impression? If it's the former, keep it flat. If it's the latter, a subtle, performance-conscious touch of depth might genuinely help, so build one small version, test it on an actual mid-range phone, and check it against reduced-motion. That single habit, depth only where impression matters, tested on real hardware, will keep you on the right side of the line between craft and gimmick.
Depth on the canvas is a real capability with a narrow set of genuinely good uses and a wide set of tempting bad ones. Use it where it serves understanding or atmosphere, keep it out of the interfaces people work in, and always pay attention to the performance and accessibility costs the demos leave out. Restraint is the whole skill.