GitHub Copilot Usage API adds review velocity metrics
The Copilot usage metrics API now includes two new fields, avg_pull_requests_minutes_to_review and avg_pull_requests_review_cycles, to measure review velocity by AI adoption phase. These metrics help quantify Copilot's impact on engineering throughput and code review efficiency for enterprise and organization reports. They are available in existing reports and apply only to merged pull requests.
- →Add review velocity metrics to Copilot usage API
- →Metrics help quantify impact of Copilot on review process
- →Metrics scoped to merged pull requests and available in reports
Features (1) ›
- Add review velocity metrics to Copilot usage API
The Copilot usage API now reports median time to first review and median review cycles for merged pull requests, broken down by AI adoption phase. These new fields, avg_pull_requests_minutes_to_review and avg_pull_requests_review_cycles, extend the existing adoption phase cohorts.
Notes (2) ›
- Metrics help quantify impact of Copilot on review process
By comparing review latency and cycle counts across adoption phases, teams can assess if deeper Copilot adoption leads to faster pull request reviews and fewer iteration cycles. This data aids in understanding Copilot's downstream effects on engineering throughput.
- Metrics scoped to merged pull requests and available in reports
The new review velocity metrics are available in both enterprise and organization 1-day and 28-day reports. They are specifically scoped to merged pull requests, ensuring they reflect work that has successfully landed.
https://github.blog/changelog/2026-07-07-add-review-cycles-and-time-to-adoption-phases-in-the-usage-api