Google Analytics AI Agent

Google Analytics AI Agent

Ask plain-language questions about your GA4 property and get back metrics, segments, and reports without opening a dashboard.

What this workflow does

1. AnalyticsQuery — type a free-text question about your site traffic
2. AnalyticsAgent — an Agent queries GA4 through the analytics-mcp MCP server and formats the answer
3. AnalyticsResult — returns the Agent's response

Use cases

• "What were my top landing pages last week?"
• "Show me traffic by country for the past 30 days"
• "Compare mobile vs desktop sessions this month vs last month"
• "Which pages have the highest bounce rate?"
• "How many new users did we get this week vs last week?"

Service account setup

1. Go to Google Cloud Console → IAM & Admin → Service Accounts
2. Create a service account (e.g. [email protected])
3. Enable the Google Analytics Data API in APIs & Services → Library
4. Download a JSON key for the service account (Actions → Manage keys → Add key → JSON)
5. In Google Analytics, go to Admin → Account Access Management → Add users, and add the service account email as a Viewer
6. Base64-encode the key: base64 < service-account.json | tr -d '\n'
7. Paste the result into the GOOGLE_APPLICATION_CREDENTIALS_JSON_B64 env field on the MCP connection

MCP connection setup

Open AnalyticsAgent, find the analytics-mcp MCP connection, and update:

• GOOGLE_APPLICATION_CREDENTIALS_JSON_B64 — your base64-encoded service account JSON
• Set your Account ID and Property ID in the system instruction (find them in GA Admin → Property Settings)

Notes

• The analytics-mcp package is installed automatically by uv run --with
• Requires Python 3.11+ and uv to be installed on the machine running Heym