What changed
GitHub expanded the Copilot usage metrics API so Copilot app activity is attributed to users and included in feature, model, language, code-activity, lines-of-code and daily-active-user rollups. The additions cover one-day and 28-day enterprise, organization and user reports.
Why it matters
Operators can compare the app with IDE, chat, review and coding-agent surfaces using the same fields instead of treating it as an opaque total. Attribution improves inventory and cost analysis, but usage and generated lines are not outcome measures.
What to do
Version the report schema, collect the same 28-day window across Copilot surfaces, reconcile user and organization totals, and pair usage with accepted changes, review rework, defects, cycle time and cost. Restrict exports and aggregate results before sharing them.
Caveat
GitHub establishes field availability and backward compatibility, not that usage causes productivity or quality. The data exposes user-level activity to privileged roles, omits inactive fields, and can be misread when generated or accepted lines are counted without downstream outcomes.