Google Analytics MCP Server for Claude

Safety notes

Google Analytics MCP Server is marked experimental upstream, so verify tool behavior and API coverage before relying on it for operational reporting.
The server uses Google Analytics Admin and Data APIs through the permissions granted to the configured credentials.
Report prompts can generate expensive or broad API queries if date ranges, dimensions, metrics, filters, or funnels are left open-ended.
Analytics results can influence marketing, product, or revenue decisions; validate important findings in Google Analytics or internal BI tools before acting.
The server is read-oriented, but it still exposes business-sensitive analytics data to the MCP client and model context.

Privacy notes

Google Application Default Credentials, OAuth client files, service account impersonation details, project IDs, and local credential paths are sensitive and should not be committed or pasted into prompts.
Account names, property IDs, Google Ads links, event names, custom dimensions, custom metrics, audiences, funnel steps, realtime activity, and report outputs can reveal business performance and user behavior.
GA4 reports can include location, device, campaign, source, medium, conversion, revenue, or audience information that may be regulated or contractually restricted.
Tool responses may be retained by MCP clients, model providers, logs, screenshots, and chat transcripts outside Google Analytics retention controls.

Prerequisites

Python 3.10 or newer with pipx available.
Google Cloud project with Google Analytics Admin API and Google Analytics Data API enabled.
Google Application Default Credentials for a user or service account with access to the intended Google Analytics accounts or properties.
OAuth consent and credential setup that includes the `https://www.googleapis.com/auth/analytics.readonly` scope.
Agreement on which GA4 properties, date ranges, dimensions, metrics, audiences, and Google Ads links an agent may inspect.

Schema details

Install type: cli
Troubleshooting: No

Source repository stats

Scope: Source repo

Collection metadata

Estimated setup: 25 minutes
Difficulty: advanced

Full copyable content

{
  "mcpServers": {
    "analytics-mcp": {
      "command": "pipx",
      "args": ["run", "analytics-mcp"],
      "env": {
        "GOOGLE_APPLICATION_CREDENTIALS": "PATH_TO_CREDENTIALS_JSON",
        "GOOGLE_PROJECT_ID": "YOUR_PROJECT_ID"
      }
    }
  }
}

About this resource

Content

Google Analytics MCP Server is an experimental local MCP server from the Google Analytics GitHub organization. It connects Claude-compatible MCP clients to the Google Analytics Admin API and Data API so they can inspect GA4 account and property metadata, list Google Ads links, retrieve custom dimensions and metrics, and run core, funnel, and realtime reports.

Use it when an agent needs current analytics context from authorized GA4 properties before summarizing traffic, campaign performance, conversions, events, funnels, or realtime activity. The server is read-oriented, but its outputs can still contain sensitive business and user-behavior data.

Source Review

These sources were reviewed on 2026-06-06. Prefer the live repository, README, PyPI metadata, license, package manifest, MCP server entrypoint, coordinator, Admin API tools, and reporting tools for current installation and behavior details.

Features

Retrieve Google Analytics account summaries and property information.
List Google Ads links for a GA4 property.
Retrieve custom dimensions and metrics.
Run GA4 core reports with the Google Analytics Data API.
Run funnel reports for defined funnel steps.
Run realtime reports for current user activity and dimensions.
Authenticate through Google Application Default Credentials.
Run locally over stdio through the analytics-mcp Python package.

Installation

Enable the Google Analytics Admin API and Google Analytics Data API in a Google Cloud project, then configure Application Default Credentials with the analytics readonly scope.

Add the server to an MCP client:

{
  "mcpServers": {
    "analytics-mcp": {
      "command": "pipx",
      "args": ["run", "analytics-mcp"],
      "env": {
        "GOOGLE_APPLICATION_CREDENTIALS": "PATH_TO_CREDENTIALS_JSON",
        "GOOGLE_PROJECT_ID": "YOUR_PROJECT_ID"
      }
    }
  }
}

Restart the MCP client and verify that analytics-mcp appears in the client's MCP server list before asking Claude to inspect accounts, properties, or reports.

Use Cases

Ask Claude to summarize available GA4 accounts and properties.
Retrieve property metadata before building a reporting prompt or dashboard.
Run traffic, event, conversion, revenue, campaign, or audience reports.
Compare funnel performance for a defined path.
Inspect realtime activity during a campaign launch or incident review.
List Google Ads links connected to a property.
Discover custom dimensions and metrics before requesting a report.

Safety and Privacy

Treat Google Analytics MCP Server as access to business analytics. Even when credentials are read-only, report outputs can reveal traffic trends, campaign performance, revenue, conversions, geography, devices, audience behavior, and other sensitive signals.

Keep OAuth client files, ADC credential files, service account details, and project IDs out of prompts and repository files. Use least-privilege accounts and only expose the GA4 properties a specific workflow needs.

Validate important business decisions against Google Analytics, your warehouse, or another approved reporting system. LLM summaries can misread date ranges, metrics, dimensions, sampling, attribution, or funnel definitions without human review.

Duplicate Check

Existing entries cover Google Workspace, Google Cloud-oriented servers, ads and marketing tools, and analytics-adjacent services, but no Google Analytics MCP Server entry, googleanalytics/google-analytics-mcp, analytics-mcp package, or matching source URL was found in content/mcp.

#google-analytics#analytics#ga4#reporting#marketing