Most engineering organizations have rolled out GitHub Copilot or another AI coding tool. The harder problem is not the rollout — it is the follow-through. Which teams are using it? Who stopped? Who has a high rejection rate that suggests a workflow mismatch? Agile Analytics 4.0 answers those questions directly inside Azure DevOps, where your team already works.
AI Metrics: Adoption at a Glance
The AI Metrics view gives any user a company-wide adoption dashboard. The Overview tab shows four headline stat cards — Active Users, Acceptance Rate, Suggestions Generated, and Teams Active — alongside a trend chart and a Setup Roadmap. The Users tab surfaces your Top 10 and Bottom 10 contributors by acceptance rate, so you know who is engaging with the tool and who has quietly stopped. The Teams tab ranks teams by adoption, giving managers and coaches a leaderboard view of organizational progress.
AIIP: From Signals to Actions
Knowing that Tina Voss has 0 active days this month is only useful if something happens next. AIIP (the AI Improvement & Intervention Panel) turns adoption signals into admin actions. Alerts are surfaced automatically — with severity ratings, the triggering reason, and one-click Acknowledge or Resolve controls. Interventions are actionable: each item links to an official Microsoft Learn or GitHub Docs resource, and admins can mark them Done when the follow-up is complete. Everything is written to an audit log.
Teams Webhook Notifications — Per Team
AIIP ships with a Microsoft Teams webhook integration. Admins configure one webhook URL per team. When an alert fires — low usage, high rejection rate, sudden drop, or milestone — the relevant team's channel gets a notification automatically. Alert types and thresholds are all configurable: set your own acceptable low-usage day count, your rejection rate ceiling, and the drop percentage that triggers an alert.
Explore Before You Connect
AI Metrics and AIIP both show realistic demo data out of the box. The full UI — stat cards, user tables, team leaderboard, alert cards, intervention items, celebration cards — is populated and explorable immediately after install, with no configuration required. When you are ready to connect a live source, go to AI Metrics → Settings and add credentials for GitHub Copilot, Microsoft 365 Copilot, or a custom endpoint. Credentials are stored in your org's ADO Extension Data Service — Microsoft-managed infrastructure, never the publisher.
AI Metrics and AIIP are available on all tiers, including Starter (free). Update to 4.0.0 from the VS Marketplace.