---
title: "Google Analytics MCP: review, setup & limitations · Undominated.ai"
canonical: https://undominated.ai/mcp-servers/google-analytics/
description: "A local connector for authorized Google Analytics property metadata and reporting APIs."
---

# Google Analytics MCP: review, setup & limitations · Undominated.ai

> A local connector for authorized Google Analytics property metadata and reporting APIs.

[← Explore all mcp servers](/mcp-servers/)

ANALYTICS AND REPORTING / Google Analytics

# Google Analytics MCP

A local connector for authorized Google Analytics property metadata and reporting APIs.

 [See setup guidance ↓](#setup)[Original source ↗](https://developers.google.com/analytics/devguides/MCP)

SOURCE REVIEW

 Reviewed 2026-10-13
 Evidence 11 linked sources
 Publisher Google Analytics
 Licence [Apache-2.0 ↗](https://github.com/googleanalytics/google-analytics-mcp/blob/75c0e1be8322e5eaa5c2e3ef7e79cf4760060f9a/LICENSE)
 Revision 75c0e1be8322
 [Read what was—and wasn’t—checked ↓](#review)

“Google Analytics MCP Server (Experimental)”

 [Google Analytics · upstream description ↗](https://github.com/googleanalytics/google-analytics-mcp/blob/75c0e1be8322e5eaa5c2e3ef7e79cf4760060f9a/README.md) Our analysis follows below.

01 / THE REASONING

## Why this made the selection.

 - Uses the publisher’s Admin and Data APIs with a read-only Analytics scope rather than scraping dashboards.
- Exposes property information, custom dimensions and historical or real-time reports for analysis within an existing account.

### A good fit for

 - Investigating traffic and event patterns in a property you can already access
- Preparing reproducible reporting questions with explicit date ranges, dimensions and metrics

### Weigh up before choosing

 - The publisher marks this server experimental; some report paths use alpha APIs and retain upstream reporting limits.
- The local process still sends authorized API requests to Google and passes report data to the assistant. It does not repair event instrumentation or prove causation.
- README project-environment examples and prose use different variable names. Verify Application Default Credentials, quota-project configuration and property access instead of relying on the sample project-ID variable.

02 / THE REVIEW RECORD

## What we actually inspected.

Source review has boundaries. A clear record is more useful than a “safe” badge.

### Material inspected

 - README.md and pyproject.toml
- Server/coordinator tool registration
- Authentication client and account/property handler
- Historical, real-time, funnel and conversion report handlers
- LICENSE notice

### Our findings

 - The credential factory requests the Analytics read-only scope and the coordinator registers reporting/metadata tools.
- The review inspected actual report calls and registration; it did not validate every reporting dimension or run an account query.
- coordinator.py registers nine tools: get_account_summaries, list_google_ads_links, get_property_details, list_property_annotations, get_custom_dimensions_and_metrics, run_report, run_realtime_report, run_funnel_report and run_conversions_report. The README lists seven of them; list_property_annotations and run_conversions_report are registered but not in the README.
- The only JSON client block in the README is the Gemini CLI ~/.gemini/settings.json mcpServers entry, which runs pipx run analytics-mcp with GOOGLE_APPLICATION_CREDENTIALS and GOOGLE_PROJECT_ID placeholders. Claude Code setup is a claude mcp add command.
- The developers.google.com MCP page, re-read on 2026-10-13, says the server is available for read requests only and cannot edit Analytics configuration or settings.

### Not established by this review

 - No MCP server or upstream package was installed or run.
- No authenticated account, API request, client handshake or write operation was tested.

The review applies to the material and revision named here. A newer upstream release can change its behavior.

03 / PUT IT TO WORK

## Connect a server deliberately.

[Upstream setup instructions ↗](https://github.com/googleanalytics/google-analytics-mcp/blob/75c0e1be8322e5eaa5c2e3ef7e79cf4760060f9a/README.md)

DOCUMENTED COMMAND

 pipx run analytics-mcp Copy command ↗

Copying does not execute this command. It may retrieve a newer version than the reviewed source.

 - Enable the Analytics Admin and Data APIs in the relevant Google Cloud project.
- Configure Application Default Credentials with Analytics read-only access and grant the identity access to the intended Analytics property.
- Configure the local MCP process using the publisher’s client-specific setup, then start with property metadata before requesting a report.
- The command above is the launch in the README’s Claude Code command (claude mcp add ... -- pipx run analytics-mcp) and in its Gemini settings.json block (command pipx, args run analytics-mcp). It was not run.

### Before you start

 - Supported Python and pipx
- Google Cloud authentication tooling or an approved equivalent ADC setup
- An identity with access to the target Analytics property

### Documented configuration example

Use the client and authentication context described in the setup guidance. Configuration formats vary between hosts.

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

### Compatibility

Gemini CLI · Claude Code · Gemini Code Assist

### Implementation

Python

### Transports

stdio

### Local MCP process with service access

 - Read account/property information and reporting data through Google APIs
- Use the configured ADC identity and project; the Analytics client requests analytics.readonly
- Expose authorized report results to the connected assistant
- Authenticates with Application Default Credentials (user OAuth client or service-account impersonation); no tool writes to Analytics.

### Cost model

Apache-2.0-licensed connector; Analytics account terms, API quotas and client/model usage remain separate.

04 / FOLLOW THE EVIDENCE

## The source trail.

Our notes are separate from the original resource. Check upstream before adopting a new version.

 - [Publisher README at reviewed revision ↗](https://github.com/googleanalytics/google-analytics-mcp/blob/75c0e1be8322e5eaa5c2e3ef7e79cf4760060f9a/README.md) Checked 2026-10-13 https://github.com/googleanalytics/google-analytics-mcp/blob/75c0e1be8322e5eaa5c2e3ef7e79cf4760060f9a/README.md Supports: summary, upstreamDescription, whySelected, bestFor, install, compatibility, limitations, access, configExample, Tool list in README
- [Source licence ↗](https://github.com/googleanalytics/google-analytics-mcp/blob/75c0e1be8322e5eaa5c2e3ef7e79cf4760060f9a/LICENSE) Checked 2026-10-13 https://github.com/googleanalytics/google-analytics-mcp/blob/75c0e1be8322e5eaa5c2e3ef7e79cf4760060f9a/LICENSE Supports: license, access.cost
- [Reviewed implementation: analytics_mcp/server.py ↗](https://github.com/googleanalytics/google-analytics-mcp/blob/75c0e1be8322e5eaa5c2e3ef7e79cf4760060f9a/analytics_mcp/server.py) Checked 2026-10-13 https://github.com/googleanalytics/google-analytics-mcp/blob/75c0e1be8322e5eaa5c2e3ef7e79cf4760060f9a/analytics_mcp/server.py Supports: review, limitations, access, transports
- [Reviewed implementation: analytics_mcp/coordinator.py ↗](https://github.com/googleanalytics/google-analytics-mcp/blob/75c0e1be8322e5eaa5c2e3ef7e79cf4760060f9a/analytics_mcp/coordinator.py) Checked 2026-10-13 https://github.com/googleanalytics/google-analytics-mcp/blob/75c0e1be8322e5eaa5c2e3ef7e79cf4760060f9a/analytics_mcp/coordinator.py Supports: review, limitations, access, transports, Nine registered tools
- [Reviewed implementation: analytics_mcp/tools/client.py ↗](https://github.com/googleanalytics/google-analytics-mcp/blob/75c0e1be8322e5eaa5c2e3ef7e79cf4760060f9a/analytics_mcp/tools/client.py) Checked 2026-10-13 https://github.com/googleanalytics/google-analytics-mcp/blob/75c0e1be8322e5eaa5c2e3ef7e79cf4760060f9a/analytics_mcp/tools/client.py Supports: review, limitations, access, transports
- [Reviewed implementation: analytics_mcp/tools/admin/info.py ↗](https://github.com/googleanalytics/google-analytics-mcp/blob/75c0e1be8322e5eaa5c2e3ef7e79cf4760060f9a/analytics_mcp/tools/admin/info.py) Checked 2026-10-13 https://github.com/googleanalytics/google-analytics-mcp/blob/75c0e1be8322e5eaa5c2e3ef7e79cf4760060f9a/analytics_mcp/tools/admin/info.py Supports: review, limitations, access, transports
- [Reviewed implementation: analytics_mcp/tools/reporting/core.py ↗](https://github.com/googleanalytics/google-analytics-mcp/blob/75c0e1be8322e5eaa5c2e3ef7e79cf4760060f9a/analytics_mcp/tools/reporting/core.py) Checked 2026-10-13 https://github.com/googleanalytics/google-analytics-mcp/blob/75c0e1be8322e5eaa5c2e3ef7e79cf4760060f9a/analytics_mcp/tools/reporting/core.py Supports: review, limitations, access, transports
- [Reviewed implementation: analytics_mcp/tools/reporting/realtime.py ↗](https://github.com/googleanalytics/google-analytics-mcp/blob/75c0e1be8322e5eaa5c2e3ef7e79cf4760060f9a/analytics_mcp/tools/reporting/realtime.py) Checked 2026-10-13 https://github.com/googleanalytics/google-analytics-mcp/blob/75c0e1be8322e5eaa5c2e3ef7e79cf4760060f9a/analytics_mcp/tools/reporting/realtime.py Supports: review, limitations, access, transports
- [Reviewed implementation: analytics_mcp/tools/reporting/funnel.py ↗](https://github.com/googleanalytics/google-analytics-mcp/blob/75c0e1be8322e5eaa5c2e3ef7e79cf4760060f9a/analytics_mcp/tools/reporting/funnel.py) Checked 2026-10-13 https://github.com/googleanalytics/google-analytics-mcp/blob/75c0e1be8322e5eaa5c2e3ef7e79cf4760060f9a/analytics_mcp/tools/reporting/funnel.py Supports: review, limitations, access, transports
- [Reviewed implementation: analytics_mcp/tools/reporting/conversions.py ↗](https://github.com/googleanalytics/google-analytics-mcp/blob/75c0e1be8322e5eaa5c2e3ef7e79cf4760060f9a/analytics_mcp/tools/reporting/conversions.py) Checked 2026-10-13 https://github.com/googleanalytics/google-analytics-mcp/blob/75c0e1be8322e5eaa5c2e3ef7e79cf4760060f9a/analytics_mcp/tools/reporting/conversions.py Supports: review, limitations, access, transports
- [Google Analytics MCP developer page ↗](https://developers.google.com/analytics/devguides/MCP) Checked 2026-10-13 https://developers.google.com/analytics/devguides/MCP Supports: homepage, Read requests only; cannot edit configuration

KEEP COMPARING

## Other approaches to consider.

Related by category or shared topics. These are alternatives to inspect, not a measured quality order.

 [### Algolia MCP Server ↗ Algolia's managed MCP offerings: Public MCP exposes application-scoped search and recommendations on selected indices for customer-facing agents; Productivity MCP gives user-scoped, read-only exploration and analytics; DevEx MCP serves version-matched API client snippets to coding agents without touching data.](/mcp-servers/algolia/)[### Amplitude MCP Server ↗ Amplitude's hosted MCP server for analysing product data, experiments and user behaviour: search dashboards, charts, notebooks and experiments, query charts and metrics, analyse experiment significance, and search session replays — with separate US and EU endpoints for data residency.](/mcp-servers/amplitude/)[### Aurora MySQL MCP ↗ AWS Labs MCP server that turns natural language into SQL for Aurora MySQL on the same host as the client, with a keyword guard the README says is not a security boundary.](/mcp-servers/awslabs-mysql/)

 [AI Tools ↗](/tools/)[Skills ↗](/skills/)[Agents ↗](/agents/)[MCP Servers ↗](/mcp-servers/)[Workflows ↗](/workflows/)

## Continue your investigation

 - [Inspect resource evidence](/tools/)
- [Choose a host](/agents/)
- [Inspect reusable workflows](/skills/)
- [Use Undominated data](/api/)
