Butter.

Exporting motion from Figma was eating my week. The plugin I relied on exported one GIF at a time, at over two minutes each, and the GIFs often came out with corrupted frames. I wrote Photoshop actions to remove them, but each design corrupted different frames, so the fix never held.

Localised versions meant detaching instances from their master just to export them, and every MP4 meant converting a GIF with another Photoshop action. A 120‑asset campaign took around six hours.

Then the plugin stopped working altogether, and came back a month later as a paid plugin. I’d already started building my own; now I needed it.

⤢ Enlarge

Myself first: I deliver 45–60 campaigns a month, some with up to 120 motion assets each. I built Butter to fix my own frustrations, but the friction of exporting motion from Figma is common. It’s designed for production designers and motion teams who need to scale localised assets without losing days to manual conversion and clean-up.

1. 01Parity first. I started with a bare-bones, Dieter Rams-style plugin that did one thing: export a single GIF at equal or better quality than the tool it replaced. Nothing else until that was right.
2. 02Then batch. With quality settled, I added exporting many assets in one go, then stress-tested it to find its limits. The hardest balance was GIF quality against file size, without letting Figma crash.
3. 03Designed in Figma, built with AI. I designed every screen and state in Figma, deciding what Butter should do in each and why, then mapped each flow and cut steps until the journey was as short as it could be. I built it with Gemini as a coding partner. It wrote the complex code; I reviewed, tested and corrected every change against my designs.
4. 04Output ready to hand off. GIFs and MP4s from the same run, compiled into one zip organised by the file’s labels or modes.

Decision

Stripped static export (PNG/JPG) prior to launch.

Trade-off

I gave up matching Figma’s own export. The static option applied to every selected asset at once, which would have confused more people than it helped. It returns as Advanced Export, designed properly this time.

⤢ Enlarge

Decision

Match quality before adding speed.

Trade-off

Batch export arrived later, but nobody would batch-export with a tool they didn’t trust for one file.

Decision

Show what will export before you export. The panel counts the selection live (2 · 1 GIF | 1 MP4).

Trade-off

A little more interface, but no surprises in the zip.

Decision

Mark MP4s with a _mp4 name suffix instead of a toggle.

Trade-off

No extra interface, and it fits how designers already name frames, but it’s hidden until you read the tip. Advanced Export replaces it with explicit per-asset controls.

Decision

Pair-programmed with an AI rather than bottlenecking the release to learn the syntax myself.

Trade-off

It shipped in months rather than years, but every change needed careful review and testing. I own the design and the logic, not every line of code, and I’m upfront about that.

• —A 120‑asset campaign now exports in 2 minutes 30 seconds instead of about six hours: a mix of GIFs and MP4s, in one zip organised by labels or modes.

⤢ Enlarge

• —No Photoshop frame fixes, no manual MP4 conversion, no detached instances.
• —Live on Figma Community since August 2026.

Advanced Export arrives at the end of Q4 2026, with granular export control: different settings per asset, directly from the panel. One frame can render as a 1× GIF and a 2× JPG while another exports as an MP4 and a 2× PNG, in the same run. Further updates are planned for 2027.