Skip to main content
Every asset tracks how it is used and reused, so you can tell which of your data people are actually finding, opening, downloading, and building on. Open an asset and select the Metrics tab in the side window.

The Metrics tab

The tab leads with the asset’s impact score — a single weighted number that rolls up everything below into one at-a-glance signal — beside a breakdown of where that impact comes from (usage, captures, reuse, citations, and social). Under it, a set of stat cards, each showing a total and a unique count:
  • Views — times the asset’s metadata was opened.
  • Downloads — times its content was downloaded or transferred out.
  • SDK / compute — programmatic accesses (the SDK pulling the asset for an automated computation), as distinct from people browsing.
  • Search impressions — times the asset appeared in a search result.
  • Sharing scope — how many distinct people can access the asset, and across how many organizations.
  • Reuse & captures — stars, the collections it belongs to, and downstream assets derived from it.
  • Citations — scholarly citations of the asset’s linked work, when present.

Total vs. unique

Each usage metric shows both a total and a unique figure. The total is every recorded action; the unique figure de-duplicates repeat activity from the same person within a short window, so it reflects how many distinct people (or machine clients) really used the asset — the more honest measure of reach. Hover the on a card to see exactly what it counts.

The impact badge

The impact score is also shown as a small donut badge you can drop next to an asset: the number sits in the center and the coloured arcs show the mix of what’s driving it. A grey ring with a dash means the asset has no recorded activity yet — distinct from a genuinely low score.

What’s counted

  • People vs. machines are tracked separately, so automated SDK/compute usage doesn’t get confused with people browsing.
  • Known web crawlers are filtered out; rapid double-clicks are collapsed — the counts follow the COUNTER “Code of Practice for Research Data” used by data repositories, so they’re comparable to views/downloads you’d see elsewhere.
  • For publications, people, and grants, public scholarly metrics (citations, field-weighted citation impact, journal impact factor) are blended in.