GA4 - Google Analytics MCP Server

What is GA4 - Google Analytics MCP Server?

GA4 - Google Analytics MCP Server is a standalone server that exposes Google Analytics 4's reporting capabilities through the standardized Model Context Protocol. It acts as a secure bridge between your GA4 property and any MCP-compatible AI tools, enabling fast analytics lookup, reporting, and integration into AI workflows through pre-defined functions.

How to Configure

1. Google Setup

   • Enable the Google Analytics Data API in your Google Cloud project.
   • Create a Service Account in "IAM & Admin > Service Accounts" and generate a JSON credentials key.
   • Grant the Service Account "Viewer" access to your GA4 property in Google Analytics.

2. Server Environment

   • Configure environment variables:
     • GOOGLE_CLIENT_EMAIL: Service account email from the credential JSON
     • GOOGLE_PRIVATE_KEY: Service account private key from the credential JSON
     • GA_PROPERTY_ID: Your GA4 Property ID

3. Installation

   • Install globally: npm install -g mcp-server-google-analytics
   • Or use with npx: npx mcp-server-google-analytics
   • Or install via Smithery for Claude Desktop integration.

4. Configuration for AI Tools (e.g., Claude Desktop)

   • Add the appropriate command and environment variables to your tool’s configuration under MCP servers.

How to Use

1. Start the Server

   • Run pnpm start, npx mcp-server-google-analytics, or launch from your tool integration.

2. Call Functions/Tools

   • AI tools or LLM agents can invoke functions such as runReport, getPageViews, getActiveUsers, getEvents, or getUserBehavior via the MCP interface.
   • Pass the required fields (e.g., date ranges, dimensions, metrics) as function arguments.

3. Example: Querying Page Views

   • Use the getPageViews tool by providing a date range and optional dimensions to get page view metrics.

4. Integration

   • Incorporate the server with any AI client or automation system that supports MCP for dynamic analytics access.

Key Features

• Access to real-time GA4 analytics data via standard MCP function calls.
• Customizable reporting: Choose specific metrics, dimensions, filters, and periods.
• Multiple pre-defined tools: Page views, user metrics, event metrics, and behavior analysis.
• Secure service account authentication, with minimal permissions needed.
• Easy installation and configuration for both standalone and integrated use (e.g., in Claude Desktop).
• Designed to be extensible for future analytics needs.

Use Cases

• Integrate GA4 metrics directly into AI-driven dashboards, reporting tools, or assistant workflows.
• Enable chatbots or AI agents to answer questions about website/app usage, traffic trends, or user engagement.
• Automate the generation of analytics summaries, cohort analysis, or event-based reports.
• Monitor campaign performance and user behavior in real-time from within AI-enabled products.

FAQ

Q1: What GA4 permissions are required for the service account?
A: The service account only needs "Viewer" access to the target GA4 property, which allows read-only access to analytics data.

Q2: Is my private key secure during operation?
A: Yes, your private key and credentials are kept only in server-side environment variables; never expose keys to the client or public repositories.

Q3: Can I use this with any AI tool or workflow?
A: Yes, as long as the tool supports MCP integration, you can connect the server and access analytics functions.

Q4: What happens if my credentials are invalid or not set up properly?
A: The server will fail to authenticate with Google Analytics, and function calls will return errors indicating credential issues; ensure you follow setup instructions closely.

Q5: Is it possible to extend or customize the available tools?
A: Yes, the project is open-source, and you can contribute or adapt the server to support additional GA4 queries or custom reporting logic.