Art, design & a little curiosity
Graphic design ↗

Figma 3D transforms: test depth before adding spectacle

Figma opened 3D transforms in beta on 12 November 2026. Use a small poster study to judge when perspective clarifies hierarchy, when it distorts reading, and what should remain flat.

Original abstract visual study for graphic design.
Original visual study / Graphic design

A new axis changes the reading order

On 12 November 2026, Figma announced that 3D transforms were available in open beta. Its release notes describe rotating layers around the X, Y, and Z axes, changing perspective and depth, and pairing scenes with motion. The help center says the effects can be applied to text, images, vectors, frames, and components; the original layer remains editable. Access requires a Full seat and edit permission. These are product statements, not evidence that a dimensional treatment will improve a particular design.

Depth adds a compositional decision: which element appears close, which recedes, and whether a shared viewpoint connects them. It can create a spatial relationship that a flat overlap cannot, but it also changes letter shapes, apparent scale, and available space. Start by naming the message and reading order. If a viewer must find a date before enjoying a title treatment, protect the date from the perspective experiment.

Separate transformation from perspective

A layer transform tilts or moves an object in space. A parent frame’s perspective establishes how that depth appears to the viewer. This distinction is useful when a design feels oddly flat: changing a layer’s depth alone may have little visible effect unless its parent keeps nested layers in 3D and provides a perspective or rotation. Figma’s documentation describes these controls and notes that a composition can instead be flattened when it needs to sit as a conventional plane inside a larger layout.

Treat the settings as a model to inspect, not as a realism switch. A strong perspective may make near objects feel larger and distant ones smaller, but it can compress copy, cause overlaps, or shift attention away from the intended focal point. Keep a flat version beside the transformed one. Compare both at the size and distance where the design will actually be read.

Make a controlled comparison

Build the smallest scene that can answer one question. For a poster, that might be a title, one image, a date, and a venue line. Duplicate the frame before experimenting so the original remains a stable reference. Change one variable at a time: first rotation, then depth, then parent perspective. Label each version with the decision it tests, such as “title turns toward image” or “date stays front-facing.”

Check the text as information, not texture. Can someone still identify the event and date in a quick glance? Does the transformed headline keep its letterforms distinguishable? Are key details obscured at any angle or during a transition? If the answer depends on decorative drama, reduce the effect or reserve it for an element that does not carry essential information. A restrained plane can give a more useful contrast to the dimensional object than making every layer participate.

Keep the handoff visible

Figma says the 3D effects map to CSS, and its developer update for the same date documents beta plugin APIs for rotation, Z translation, perspective, and motion keyframes. That connection is a handoff aid, not a guarantee that every destination, device, or implementation will look identical. Inspect the copied transform in context, and decide how the design should behave when motion is unavailable, reduced, or not yet implemented. A useful design file makes the intended hierarchy legible even before the effect runs.

Put it into practice

Try it yourself

  1. Create a compact poster frame with a headline, image, date, and venue. Duplicate it before applying effects.
  2. Make three variants: flat, rotated layers with subtle depth, and the same layers under a stronger shared perspective. Change one setting at a time and label the question each variant answers.
  3. Preview the headline and essential details at thumbnail size, then inspect the CSS handoff. Record one benefit, one reading cost, and one condition that would make you remove the effect.
A worked example

Worked example: a fictional planetarium poster

Imagine a poster for an invented evening program called “Moon Maps.” Keep the date and venue in a flat footer. In the first study, place the title straight across a dark sky image. In the second, tilt the title slightly around its vertical axis while a circular moon diagram sits farther back in the same parent frame. In the third, exaggerate the perspective to see where the letterforms begin to feel cramped. Compare the three at a small phone-card size: if “Moon Maps” reads but the date disappears, adjust the layout rather than shrinking the date to rescue the effect. The program and poster are fictional; this exercise reports no Figma test or audience result.

References & further reading

Examples and exercises in this guide are original teaching scenarios.

Keep following the thread.

More graphic design ↗
Graphic design4 min read

Firefly Design beta: test the workflow, not the promise

Adobe announced Firefly Design beta at MAX on 10 November 2026, bringing vector, raster, motion, and 3D work into one canvas. Use a small, repeatable design task to judge whether a unified workspace helps your process and where your own review still matters.