Google Analytics 4 MCP server

The Google Analytics MCP Server provides a Model Context Protocol integration for querying Google Analytics 4 data using natural language. This tool allows you to connect your GA4 property to Claude, Cursor, and other MCP clients, enabling access to over 200 dimensions and metrics for analyzing website traffic, user behavior, and analytics data.

Prerequisites

Before installing the Google Analytics MCP server, ensure you have:

• Python 3.10 or higher
• A Google Analytics 4 property with data
• A service account with Analytics Reporting API access

You can verify your Python setup with:

# Check Python version (need 3.10+)
python --version
python3 --version

# Check pip
pip --version
pip3 --version

Setting Up Google Analytics Credentials

Create a Service Account in Google Cloud Console

1. Go to Google Cloud Console
2. Create or select a project:
   • New project: Click "New Project" → Enter project name → Create
   • Existing project: Select from dropdown
3. Enable the Analytics APIs:
   • Go to "APIs & Services" → "Library"
   • Search for "Google Analytics Data API" → Click "Enable"
4. Create Service Account:
   • Go to "APIs & Services" → "Credentials"
   • Click "Create Credentials" → "Service Account"
   • Enter name (e.g., "ga4-mcp-server")
   • Click "Create and Continue"
   • Skip role assignment → Click "Done"
5. Download JSON Key:
   • Click your service account
   • Go to "Keys" tab → "Add Key" → "Create New Key"
   • Select "JSON" → Click "Create"
   • Save the JSON file - you'll need its path

Add Service Account to GA4

1. Get service account email:
   • Open the JSON file
   • Find the client_email field
   • Copy the email (format: [email protected])
2. Add to GA4 property:
   • Go to Google Analytics
   • Select your GA4 property
   • Click "Admin" (gear icon at bottom left)
   • Under "Property" → Click "Property access management"
   • Click "+" → "Add users"
   • Paste the service account email
   • Select "Viewer" role
   • Uncheck "Notify new users by email"
   • Click "Add"

Find Your GA4 Property ID

1. In Google Analytics, select your property
2. Click "Admin" (gear icon)
3. Under "Property" → Click "Property details"
4. Copy the Property ID (numeric, e.g., 123456789)
   • Note: This is different from the "Measurement ID" (starts with G-)

Test Your Setup (Optional)

Verify your credentials:

pip install google-analytics-data

Create a test script (test_ga4.py):

import os
from google.analytics.data_v1beta import BetaAnalyticsDataClient

# Set credentials path
os.environ["GOOGLE_APPLICATION_CREDENTIALS"] = "/path/to/your/service-account-key.json"

# Test connection
client = BetaAnalyticsDataClient()
print("✅ GA4 credentials working!")

Run the test:

python test_ga4.py

Installing the MCP Server

Choose one of these installation methods:

Method A: pip install (Recommended)

pip install google-analytics-mcp

MCP Configuration:

First, check your Python command:

python3 --version
python --version

Then use the appropriate configuration:

If python3 --version worked:

{
  "mcpServers": {
    "ga4-analytics": {
      "command": "python3",
      "args": ["-m", "ga4_mcp_server"],
      "env": {
        "GOOGLE_APPLICATION_CREDENTIALS": "/path/to/your/service-account-key.json",
        "GA4_PROPERTY_ID": "123456789"
      }
    }
  }
}

If python --version worked:

{
  "mcpServers": {
    "ga4-analytics": {
      "command": "python",
      "args": ["-m", "ga4_mcp_server"],
      "env": {
        "GOOGLE_APPLICATION_CREDENTIALS": "/path/to/your/service-account-key.json",
        "GA4_PROPERTY_ID": "123456789"
      }
    }
  }
}

Method B: GitHub download

git clone https://github.com/surendranb/google-analytics-mcp.git
cd google-analytics-mcp
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

MCP Configuration:

{
  "mcpServers": {
    "ga4-analytics": {
      "command": "/full/path/to/ga4-mcp-server/venv/bin/python",
      "args": ["/full/path/to/ga4-mcp-server/ga4_mcp_server.py"],
      "env": {
        "GOOGLE_APPLICATION_CREDENTIALS": "/path/to/your/service-account-key.json",
        "GA4_PROPERTY_ID": "123456789"
      }
    }
  }
}

Updating Configuration

Replace these placeholders in your MCP configuration:

• /path/to/your/service-account-key.json with your JSON file path
• 123456789 with your GA4 Property ID
• /full/path/to/ga4-mcp-server/ with your download path (Method B only)

Usage

Once configured, you can ask your MCP client various questions:

Discovery & Exploration

• What GA4 dimension categories are available?
• Show me all ecommerce metrics
• What dimensions can I use for geographic analysis?

Traffic Analysis

• What's my website traffic for the past week?
• Show me user metrics by city for last month
• Compare bounce rates between different date ranges

Multi-Dimensional Analysis

• Show me revenue by country and device category for last 30 days
• Analyze sessions and conversions by campaign and source/medium
• Compare user engagement across different page paths and traffic sources

E-commerce Analysis

• What are my top-performing products by revenue?
• Show me conversion rates by traffic source and device type
• Analyze purchase behavior by user demographics

Quick Start Examples

Try these example queries:

1. Geographic Distribution

Show me a map of visitors by city for the last 30 days, with a breakdown of new vs returning users

2. User Behavior Analysis

Compare average session duration and pages per session by device category and browser over the last 90 days

3. Traffic Source Performance

Show me conversion rates and revenue by traffic source and campaign, comparing last 30 days vs previous 30 days

4. Content Performance

What are my top 10 pages by engagement rate, and how has their performance changed over the last 3 months?

Performance Optimizations

This MCP server includes built-in optimizations:

Smart Data Volume Management

• Automatic row estimation - Checks data volume before fetching
• Interactive warnings - Alerts when queries would return >2,500 rows
• Optimization suggestions - Provides specific recommendations to reduce data volume

Server-Side Processing

• Intelligent aggregation - Automatically aggregates data when beneficial
• Smart sorting - Returns most relevant data first
• Efficient filtering - Leverages GA4's server-side filtering capabilities

User Control Parameters

• limit - Set maximum number of rows to return
• proceed_with_large_dataset=True - Override warnings for large datasets
• enable_aggregation=False - Disable automatic aggregation
• estimate_only=True - Get row count estimates without fetching data

Available Tools

The server provides 5 main tools:

1. get_ga4_data - Retrieve GA4 data with custom dimensions and metrics
2. list_dimension_categories - Browse available dimension categories
3. list_metric_categories - Browse available metric categories
4. get_dimensions_by_category - Get dimensions for a specific category
5. get_metrics_by_category - Get metrics for a specific category

Dimensions & Metrics

Access to over 200 GA4 dimensions and metrics organized by category:

Dimension Categories

• Time: date, hour, month, year, etc.
• Geography: country, city, region
• Technology: browser, device, operating system
• Traffic Source: campaign, source, medium, channel groups
• Content: page paths, titles, content groups
• E-commerce: item details, transaction info
• User Demographics: age, gender, language
• Google Ads: campaign, ad group, keyword data
• And 10+ more categories

Metric Categories

• User Metrics: totalUsers, newUsers, activeUsers
• Session Metrics: sessions, bounceRate, engagementRate
• E-commerce: totalRevenue, transactions, conversions
• Events: eventCount, conversions, event values
• Advertising: adRevenue, returnOnAdSpend
• And more specialized metrics

Troubleshooting

If you get "No module named ga4_mcp_server" (Method A):

pip3 install --user google-analytics-mcp

If you get "executable file not found":

• Try the other Python command (python vs python3)
• Use pip3 instead of pip if needed

Permission errors:

# Try user install instead of system-wide
pip install --user google-analytics-mcp

Credentials not working:

1. Verify the JSON file path is correct and accessible
2. Check service account permissions:
   • Go to Google Cloud Console → IAM & Admin → IAM
   • Find your service account → Check permissions
3. Verify GA4 access:
   • GA4 → Admin → Property access management
   • Check for your service account email
4. Verify ID type:
   • Property ID: numeric (e.g., 123456789) ✅
   • Measurement ID: starts with G- (e.g., G-XXXXXXXXXX) ❌

API quota/rate limit errors:

• GA4 has daily quotas and rate limits
• Try reducing the date range in your queries
• Wait a few minutes between large requests

How to install this MCP server

For Claude Code

To add this MCP server to Claude Code, run this command in your terminal:

claude mcp add-json "ga4-analytics" '{"command":"python3","args":["-m","ga4_mcp_server"],"env":{"GOOGLE_APPLICATION_CREDENTIALS":"/path/to/your/service-account-key.json","GA4_PROPERTY_ID":"123456789"}}'

See the official Claude Code MCP documentation for more details.

For Cursor

There are two ways to add an MCP server to Cursor. The most common way is to add the server globally in the ~/.cursor/mcp.json file so that it is available in all of your projects.

If you only need the server in a single project, you can add it to the project instead by creating or adding it to the .cursor/mcp.json file.

Adding an MCP server to Cursor globally

To add a global MCP server go to Cursor Settings > Tools & Integrations and click "New MCP Server".

When you click that button the ~/.cursor/mcp.json file will be opened and you can add your server like this:

{
    "mcpServers": {
        "ga4-analytics": {
            "command": "python3",
            "args": [
                "-m",
                "ga4_mcp_server"
            ],
            "env": {
                "GOOGLE_APPLICATION_CREDENTIALS": "/path/to/your/service-account-key.json",
                "GA4_PROPERTY_ID": "123456789"
            }
        }
    }
}

Adding an MCP server to a project

To add an MCP server to a project you can create a new .cursor/mcp.json file or add it to the existing one. This will look exactly the same as the global MCP server example above.

How to use the MCP server

Once the server is installed, you might need to head back to Settings > MCP and click the refresh button.

The Cursor agent will then be able to see the available tools the added MCP server has available and will call them when it needs to.

You can also explicitly ask the agent to use the tool by mentioning the tool name and describing what the function does.

For Claude Desktop

To add this MCP server to Claude Desktop:

1. Find your configuration file:

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json
• Linux: ~/.config/Claude/claude_desktop_config.json

2. Add this to your configuration file:

{
    "mcpServers": {
        "ga4-analytics": {
            "command": "python3",
            "args": [
                "-m",
                "ga4_mcp_server"
            ],
            "env": {
                "GOOGLE_APPLICATION_CREDENTIALS": "/path/to/your/service-account-key.json",
                "GA4_PROPERTY_ID": "123456789"
            }
        }
    }
}

3. Restart Claude Desktop for the changes to take effect