• 4 min read • AI Data Stream Team

GitHub Is Now a Connected Data Source

Connect your GitHub repository to AI Data Stream. Correlate deployments, releases, and commits with your analytics data. Available on Pro and Business plans.

update product analytics

Traffic dropped 15% overnight. Was it an algorithm update? A campaign ending? Or did someone deploy a breaking change at 5pm on Friday?

Until now, answering that last question meant switching to GitHub, scanning commit history, cross-referencing timestamps, and hoping you could piece it together. Now the AI can just check for you.

GitHub is now available as a connected data source on Pro and Business plans.

What This Means for You

When you connect a GitHub repository, the AI gains access to your development timeline - releases, commits, deployments, issues, and pull requests. The real value is what happens when you combine this with your analytics data.

Correlate Deployments with Traffic Changes

This is the big one. Ask the AI “why did traffic drop last Thursday?” and it can now check whether a deployment or release happened around that time, what code changed, and whether any related issues were reported.

No more tab-switching between GA4 and GitHub trying to match timestamps.

Track Releases Against Performance

Launched a new feature? The AI can tell you how traffic, conversions, and engagement shifted after the release. Ask questions like:

  • “How did traffic change after the v2.0 release?”
  • “Compare conversion rates before and after last week’s deployment”
  • “Did the homepage redesign affect bounce rate?”

Cross-Reference Bugs with Metrics

Got open issues labelled “bug” or “performance”? The AI can connect the dots between known issues and metric changes. If bounce rate spiked on mobile and there’s an open issue about mobile layout breaking, the AI will surface that connection.

Monitor Development Velocity

Track commit frequency and deployment cadence alongside your growth metrics. Useful for agencies managing multiple client sites, or teams that want to understand the relationship between shipping speed and business outcomes.

See It on Your Dashboard

GitHub gets its own dashboard widget showing metric cards for commits, releases, open PRs, and open issues, plus a recent activity feed with links back to GitHub. Like all dashboard widgets, it’s fully draggable and resizable in edit mode.

But the feature we’re most excited about is the traffic chart overlay. Toggle “GitHub events” on your GA4 traffic chart and you’ll see:

  • Release markers (dark triangles) on each release date
  • Commit markers (purple circles) sized by the number of commits that day

Hover over any marker for details. It makes deployment correlation visual and instant — no questions needed, just look at the chart.

The overlay respects your per-tool settings, so if you’ve disabled commits in your connection config, they won’t clutter the chart.

What Data the AI Can Access

When you connect a repository, seven tools become available:

  • Repository overview - Language, stars, forks, visibility, license
  • Releases - Tag names, dates, release notes
  • Recent commits - Messages, authors, dates (filterable by branch, date range, or author)
  • Commit activity - Weekly commit frequency over the last year
  • Deployments - Production and staging deployments with status
  • Issues - Open and closed issues with labels and assignees
  • Pull requests - Merged, open, and closed PRs with branch info

Each tool can be individually enabled or disabled in your connection settings, so you only expose what you need.

All access is read-only. We never modify your repository.

How to Connect

You have two options:

OAuth (recommended): Click Add Connection, select GitHub, and authorize through GitHub’s OAuth flow. Takes about 30 seconds.

Personal Access Token: Generate a token at github.com/settings/tokens with the repo scope, then paste it into AI Data Stream. Better for fine-grained access control.

Both methods work identically once connected. Full setup instructions are in our Connecting Data Sources documentation.

Availability

GitHub is available on Pro and Business plans, sitting alongside Google Ads and WordPress at the Pro tier. If you are already on Pro or Business, you can connect a repository right now from your property’s Connections page.

On the free plan? Upgrade to Pro to get GitHub along with Google Ads, WordPress, and unlimited properties.

Example Questions to Try

Once connected, try asking your AI:

What releases were deployed in the last 30 days?

Did any deployments coincide with the traffic drop last week?

Show me commit activity trends over the last 3 months.

Which pull requests were merged this week?

Are there open issues that could explain the conversion decrease?

Compare traffic before and after the last production deployment.

The AI works best when you combine GitHub data with your other connections. A question like “did a deployment cause the traffic drop?” pulls from both GitHub and GA4 to give you an answer with full context.

Get Started

Head to your property’s Connections page and add GitHub. If you have questions, check the full documentation or contact us.

Related Posts

Why We Don't Charge for AI (And What That Means for You)

Chat with your Google Analytics, Search Console, and Google Ads data using your own AI. No markup on API costs, no usage limits, no surprise price hikes.

Keeping Your AI Analytics Costs Low

With BYOK pricing, you control your AI costs directly. Here are practical ways to keep them minimal - from picking the right model to managing conversation context.

How System Prompts Work - And Why They Matter for AI Analytics

The system prompt is the difference between an AI that hallucinates plausible nonsense and one that stays grounded in your actual data. Here's how they work, and what good analytics prompts look like.