How to Connect GitHub to Power BI - Track Repos & Commits Easily 2026

Everything begins in My workspace, the default landing page in the Power BI service where your reports, semantic models, and dashboards are listed by name, type, owner, and last refresh time. Before connecting any new source, it's worth glancing at what's already here — this is also where the GitHub-fed report will eventually live once it's built, alongside whatever other reports and dashboards your workspace already has.

Since there's no native GitHub connector, authentication runs through GitHub's own token system rather than an OAuth prompt inside Power BI. In your GitHub account settings, create a personal access token scoped for read access to repositories and commits — for an organization repo, make sure the token has the right org-level permissions too, since a token that only covers your personal repos will fail silently against anything you don't directly own.

From the workspace, the Apps section opens a marketplace of installable apps from Microsoft and third parties, covering everything from inventory and emissions dashboards to finance and manufacturing templates. It's worth searching this list before building a connection from scratch, since a pre-built template app can save you the work of designing the report layer yourself — though for GitHub specifically, most teams end up building their own connection rather than finding an exact match here.

Back in Power BI Desktop, use Get Data and choose Web, then point it at a GitHub REST API endpoint for the repository you want — commits, issues, or pull request data all have their own endpoints. Paste your personal access token into the HTTP request header when prompted for authentication rather than the URL itself, since embedding it in the address is both less secure and more likely to break if GitHub rotates its API format.

GitHub's API returns commit and repo data as nested JSON, which Power Query needs to expand into flat columns before it's usable in a visual — expect to click through a few Expand and Convert to Table steps to get commit author, date, and message into their own columns. Filter out any noise, like automated bot commits or merge commits, if they'd otherwise skew a chart of contributor activity.

A line chart of commits over time by contributor, or a table comparing commit counts across repositories, are the two views most teams reach for first. Publish the report back to the workspace you started in, then set a scheduled refresh — GitHub API rate limits mean an hourly refresh is usually more than sufficient, and a refresh scheduled too aggressively risks hitting those limits for no real benefit.