I’m in a large org with a mature design system (~75 components), pulling library usage data from the Figma REST API.

When I sort the output by component, the API spits out every variant and sub component in the library. This means that a design system library with 75 components might show as several hundred components in the data… which is too messy to be useful for my stakeholders.

How are people filtering down this data to make it readable?

In my pilots I categorized components using a manually populated table, but that’s not an approach that is scalable (when I scan all 10 version of our design system library I wind up with thousands of components to categorize).

You can see in the Library Analytics Modal within Figma that they group variants under their main components (I’m assuming using the file hierarchy to group them) but this isn’t available in the API.