Figma Make Is Actually 20% Faster — Here's What the Numbers Mean for You

Figma's data science team ran a randomized controlled trial — the same rigorous research standard used in clinical studies — to measure exactly how much time Figma Make saves. The headline numbers: design work got 20% faster and 16% easier overall, with product managers seeing even bigger gains (23% faster, 37% easier). That's not a vibe. That's a controlled experiment with 100 participants.

This matters because "AI saves time" has become white noise. What's different here is that Figma actually tried to prove it — and the methodology they chose tells you something about how seriously to take the result.

Why the Research Method Actually Matters

Most AI productivity claims come from surveys or from comparing users who chose to use AI against users who didn't. The problem with that approach: people who reach for AI tools are probably already faster, more comfortable with new tools, or working on simpler tasks. You can't untangle "AI helped" from "that person was already efficient."

Figma's team considered the standard approaches — A/B tests and log-data analysis — and rejected both, because neither could adequately control for confounders (things like a participant's years of experience or the complexity of a given task that would skew results regardless of AI). Instead they ran an RCT: a randomized controlled trial, where 100 participants were split into two groups and asked to complete the exact same tasks — one group with Figma Make, one without. Same tasks, randomized groups, trained moderators. That structure is why the 20% figure is worth paying attention to, rather than filing under "vendor marketing."

What the Results Actually Say

The study found that designers using Make completed tasks 20% faster and rated the work 16% easier. Product managers — who aren't typically Figma power users — saw the sharpest improvement: 23% faster and 37% easier.

The PM result is the more interesting one. Figma has said that Make has opened the door to design for PMs — meaning people who think in requirements and user stories, not components and constraints. A near-40% ease improvement suggests Make is genuinely lowering the floor, not just speeding up people who were already fluent in the tool.

For designers who are just starting to build their own apps with AI tools, that's the signal worth holding onto: Make appears to compress the gap between "I have an idea" and "I have something I can test."

How to Actually Use This as a Designer-Who-Builds

If you're using Figma Make (or exploring it), here's how to think about putting this research to work:

Use Make for the mechanical middle. The study was specifically designed around everyday design tasks — the repetitive, procedural work that slows a project down. That's where Make earns its keep. Generating initial UI from a prompt, exploring layout variations, scaffolding an app structure — these are the moments to reach for it, not the nuanced decisions about hierarchy or user flow that still need your judgment.

Don't skip the prototype phase. Make's strength is turning a prompt into something tangible quickly. As a designer-builder, your advantage is that you can evaluate what it produces with a trained eye — and iterate faster than a developer-only team. Use that loop deliberately: prompt, critique like a designer, refine, repeat.

If you're a PM reading this: the data suggests Make is genuinely built for you too, not just as a spectator tool. Exploring it to mock up a feature before handing it to a design team could meaningfully change how early conversations go.

What the Study Doesn't Tell Us

A 100-person controlled trial is rigorous by tech-industry standards, but it's still a snapshot. The tasks were standardized — which is what makes the results clean — but your actual workday is messier. Complex, novel problems with ambiguous briefs may not mirror the conditions of the study.

It's also worth noting that Figma designed and funded this research. The methodology is transparent and the RCT approach is sound, but independent replication would make the numbers more definitive. The full study is available to read, which is more than most companies offer — take a look if you want to stress-test the methodology yourself.

The Takeaway

A 20% time saving, validated with real experimental controls, is meaningful — not transformative overnight, but meaningfully real. For a designer learning to build, the more important signal is the PM result: Make is compressing the distance between non-technical intent and working prototype. That's the same gap you're probably trying to close. The tool won't replace design thinking, but it does seem to genuinely handle the mechanical work that used to eat your afternoon.