Skip to main content

Workspace-level metric definitions, managed by admins

A metric catalog at workspace level. Admins define each metric once, and every dataset, destination and AI query reuses it.

Each metric definition holds:

  • Name and description. For example, blended_cpa: “Ad spend across Google, Meta and LinkedIn ÷ CRM deals created”.

  • Formula. How to calculate the metric from its parts (for example, total spend ÷ total deals). Coupler.io computes it for whatever period or breakdown is asked for, so totals and averages are always right.

  • Unit and currency.

  • Default filters and exclusions. For example, “exclude campaigns containing ‘test’”.

  • Owner, last edited by, and version history.

How it behaves:

  • Permissions. Admins create and edit metrics; members can only use them. Optionally, members could propose a change for an admin to approve.

  • Binding to datasets. A metric binds to datasets by column mapping (spend → cost, deals → deals_created), so one definition works across differently shaped datasets.

  • MCP and AI. The AI sees the metric catalog and calls metrics by name, and Coupler.io computes them. Answers cite the metric name and its version.

  • Change history. When a definition changes, history shows from which date each version applied.

Log in to comment and vote

No comments yet

Be the first to share your thoughts.