SEO Cartograph · MCP

Google Analytics MCP: plug GA4 into your AI

Q: What is a Google Analytics MCP server?

An MCP (Model Context Protocol) server that exposes your Google Analytics 4 data to an AI such as Claude. Instead of exporting a GA4 report and pasting it into a chat, the AI queries the property directly: sessions, users, conversions and revenue, by page, channel and country, over whatever date range you ask for. SEO Cartograph exposes GA4 and Search Console through the same connector, so traffic and search performance can be cross-checked in a single question.

Your AI queries your Google Analytics 4 property directly: sessions, users, conversions and revenue, by page, channel and country, over whatever date range you ask for. No export, no copy-paste, and on the same data as your Search Console, through one connector.

Get started Pricing

Claude · SEO Cartograph MCP

Which landing pages lost conversions this quarter, and where was their traffic coming from?

Tools called

explore_analytics get_analytics_history

dimension landingPage · metrics traffic · 90 daysread only

The AI reads GA4, cross-checks Search Console, you decide

The protocol

What is a Google Analytics MCP server?

MCP (Model Context Protocol) is the open standard that lets an AI read data and call tools inside a third-party application. SEO Cartograph exposes GA4 through that channel: your AI phrases a question in plain language, the tool turns it into an Analytics API call, and the rows come back. One connector covers GA4, Search Console, the crawl and rank tracking.

The data

What your AI reads in GA4

Exploration is open-ended: you pick the dimension, the date range and the metric set, and the AI composes the query.

Dimensions

landingPage
The entry page, the one that matters for SEO.
pagePath
Every page viewed, down to the path.
sessionDefaultChannelGroup
Channel: organic, direct, referral, paid, social.
sessionSourceMedium
Source and medium, in full detail.
country, city, deviceCategory
Country, city, device type.
eventName, newVsReturning, date
Events, new versus returning, day by day.

Metric sets

traffic
Sessions, users, conversions, engagement, bounce rate, duration, revenue.
events
Event counts and users per event.
products
Purchases, revenue, product views and add-to-carts, for ecommerce.

Filters

Any date range, defaulting to the last 28 days.
A dimension-equals-value filter, organic channel only for instance.
Up to 5,000 rows per call, enough for a large catalogue.

AI-engine traffic, isolated

The Analytics exploration tool accepts a mode that keeps only sessions coming from ChatGPT, Perplexity, Gemini, Claude or Copilot. Combined with source, you get sessions engine by engine. Combined with landing page, the pages they actually send people to. That measurement is missing from most dashboards, and here it comes out of a single question.

The tools

The two GA4 tools in the connector

The badge shows the plan required for Analytics data.

`explore_analytics` Startup

Open-ended navigation inside the property. Any dimension, any range, with or without a filter, and three metric sets. This is the tool for open questions: traffic by channel, conversions by landing page, sales by product, sessions by country.

`get_analytics_history` Startup

How the site moves over time, weekly and monthly, with daily figures over 90 days for fine grain. GA4 sessions and conversions arrive already lined up with Search Console clicks and impressions, and with the position breakdown.

Why it matters

GA4 and Search Console in the same question

A GA4-only connector answers “how many sessions”. Combined with Search Console, it answers “why”.

A page loses sessions: the AI checks whether impressions dropped, whether the position slipped, or whether only the CTR got worse.
A page gains clicks without gaining conversions: the issue sits on the page itself.
A channel falls off: the weekly history pins down the week it turned, and the crawl says whether a technical change landed at the same time.
AI-engine traffic climbs: you see which pages carry it, and you pick your next topics from there.

Setup

Plugging GA4 into your AI, in 3 steps

The Google Analytics connector is set up once in the dashboard, then the AI hooks into it.

Connect Google Analytics to the project

In the dashboard, using the Google account with access to the property.

Install the SEO Cartograph connector

One command through the Claude marketplace, or the MCP server URL for any other compatible client.

Ask your question

In plain language. The AI picks the dimension and the range, and hands you the rows.

Installation

MCP server URL

https://app.tnedjar.com/mcp/cartograph

Install through the Claude marketplace

claude plugin marketplace add tom-ned/seo-cartograph-mcp

claude plugin install seo-cartograph

In Claude Desktop: Settings, Plugins, Add a marketplace

Day to day

Questions worth asking once you are connected

Copy them straight into your MCP client.

Prompts

Compare organic sessions this quarter against last, by landing page, and give me the ten biggest losses.

Which pages get traffic from AI engines, and which of them convert?

Cross-check conversions by landing page with the Search Console queries for those same pages.

Over the last 90 days, which channel has the best engagement rate, and in which countries?

Which products lost revenue this month, and did traffic to their pages move?

The same questions work in Claude.ai, Claude Desktop and Claude Code

Limits

What the GA4 connector does not do

Worth saying before you install it.

No writes to GA4

Access is read-only. No conversion, audience or filter is created, and no property setting is changed.

Your GA4 settings rule

If the property keeps two months of data, the AI will not see fourteen. Retention is set in Google Analytics, not here.

Fresh to yesterday

GA4 consolidates the previous day. For live figures, Google’s realtime report remains the place to look.

Your data stays yours

The point that matters when you plug an AI into a client’s data.

Read-only on Analytics. The AI reads; it writes nothing into your property.
OAuth access with a consent screen. You see what the connector can read before you authorise it.
Data scoped to your account. The connector exposes only the projects in your workspace.
Revocable in one click. From your MCP client or from your workspace.

The intelligence is paid for by your Claude subscription

No AI surcharge on our side, no token meter, no credits to top up. You plug in the AI you already pay for, and SEO Cartograph supplies the data and the tools. The only meter that moves on our side covers live SERP analyses.

FAQ

Frequently asked questions

What is a Google Analytics MCP server?

An MCP server that exposes your Google Analytics 4 data to an AI. Instead of exporting a GA4 report and pasting it into a chat, the AI queries the property directly: sessions, users, conversions and revenue, by page, channel and country, over whatever date range you ask for. SEO Cartograph exposes GA4 and Search Console through the same connector, so traffic and search performance can be cross-checked in a single question.

Which plan do I need to connect GA4 to my AI?

The MCP connector is available from the free plan, and Analytics data opens up on the Startup plan, at 15.-/month, with six months of history. Croissance, at 49.-/month, moves to full history. The pricing page compares all four plans. On the AI side you need a Claude plan that allows connectors.

Can the AI change my Google Analytics property?

No. GA4 access is read-only. The AI reads your reports; it creates no conversions, audiences or filters, and changes no property setting.

Can I see traffic coming from ChatGPT and other AI engines?

Yes. The Analytics exploration tool accepts a mode that keeps only sessions coming from AI engines: ChatGPT, Perplexity, Gemini, Claude or Copilot. Combined with the source dimension it gives you sessions engine by engine, and combined with landing page, the pages they actually send people to.

How far back can the AI look?

Over whatever date range you ask for, within what your property retains. GA4 applies an adjustable retention period, and 14 months is the setting to pick if you want year-on-year comparisons. Data is fresh to yesterday.

Are there extra AI credits to pay for?

No. The intelligence comes from the Claude subscription you already pay for. SEO Cartograph supplies the data and the tools, with no token meter and no AI credits to top up.

Which MCP clients are supported?

The connector follows the MCP standard. It is built and tested for Claude.ai, Claude Desktop and Claude Code, and any protocol-compliant client can connect through the server URL. The MCP connector page covers the setup and the full tool list.

Why plug GA4 into an AI rather than read the reports

A Google Analytics report answers the question it was built to ask. As soon as the question changes you add a dimension, change the range, apply a filter, and start over. With an MCP connector the question is one sentence and the tool composes the query. The difference shows up most on multi-storey questions, the ones where you look at conversions by landing page, then go back to the Search Console queries for those same pages, then check what the crawl says about their technical state. Three interfaces normally, one conversation here.

What it changes for an agency

A client’s monthly review fits into one question, and the figures come from the client’s own property. The AI pulls out how traffic moved, pins down the week a channel fell off, and lines that date up against work done on the site. The same connector then feeds internal linking proposals and on-page fixes, with the GSC and GA4 client dashboard as the reporting surface.

One connector, several sources

GA4 is one source among several. The same MCP server reaches Search Console, the technical crawl, semantic clusters and rank tracking. That is what lets a traffic question get an SEO answer, without switching tools or gluing exports back together. The features page covers the full scope, and the SEO MCP page lists the tools.