{"id":1827,"date":"2026-02-26T10:06:19","date_gmt":"2026-02-26T10:06:19","guid":{"rendered":"https:\/\/devblogs.microsoft.com\/all-things-azure\/?p=1827"},"modified":"2026-02-26T10:06:19","modified_gmt":"2026-02-26T10:06:19","slug":"measuring-actual-ai-impact-for-engineering-with-apache-devlake","status":"publish","type":"post","link":"https:\/\/devblogs.microsoft.com\/all-things-azure\/measuring-actual-ai-impact-for-engineering-with-apache-devlake\/","title":{"rendered":"Measuring actual AI Impact for Engineering with Apache DevLake"},"content":{"rendered":"

If you want to skip the explain and get started super quick with adoption + impact insights, use gh-devlake<\/a> to deploy a GitHub Copilot impact dashboard in a few CLI commands.<\/em><\/p>\n

\n

So! You’ve rolled out GitHub Copilot to your engineering teams. You’ve got the built-in dashboards. You know how many seats are assigned, what the acceptance rates look like, which editors your teams prefer. Maybe you’ve even pulled the Copilot Metrics API and built some charts.<\/p>\n

But here’s the question your VP of Engineering or CTO is actually asking:<\/p>\n

“Is GitHub Copilot making us ship faster? Are we more reliable? Is code review getting better?”<\/p><\/blockquote>\n

And the honest answer, if you’re only looking at data from our Copilot Metrics API, is: you don’t know<\/em>. You know adoption. You don’t know impact.<\/p>\n

I’ve been working with engineering teams evaluating Copilot for a while now, and this gap \u2014 between usage data and delivery outcomes \u2014 keeps coming up. The usage metrics live in one place. The deployment data lives in your CI\/CD tool. The incident data lives in your issue tracker. The code review data lives in GitHub. They’re all siloed. And the question leadership is asking requires connecting them.<\/p>\n

This post is about a solution that connects them:\u00a0Apache DevLake<\/a>. And more specifically, why it’s\u00a0not<\/em>\u00a0just another dashboard \u2014 it’s a data warehouse that lets you actually answer the impact question.<\/p>\n

\"Apache<\/a><\/p>\n

 <\/p>\n

\n

The Observability Gap<\/h2>\n

GitHub gives you solid Copilot metrics out of the box<\/a>:<\/p>\n