Integrates with Google Ads, supporting raw GAQL queries, campaign performance, keyword performance with quality score, ad group metrics, and account summary with top campaigns.
Provides tools for interacting with Google Analytics 4, including listing accounts, running custom reports, retrieving real-time data, top pages, traffic sources, and user metrics with period-over-period comparison.
Offers tools for Google Search Console, such as listing sites, executing search analytics queries, retrieving top queries and top pages, inspecting URL indexing status, and listing sitemaps.
Enables access to Meta Ads data, with tools to list ad accounts, view account-level spend/performance, and drill into campaign, ad set, and individual ad metrics.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@mcp-marketing-analytics Show me my top pages for the last 30 days"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
A Model Context Protocol (MCP) server that gives Claude access to Google Analytics 4, Google Search Console, Meta Ads, and Google Ads. Ask Claude about your marketing data in natural language.
Tool | Description |
| Step-by-step setup instructions — shows what's configured and guides you through any missing API credentials |
Tool | Description |
| List all GA4 properties accessible to your Google account |
| Run a custom report with any dimensions/metrics |
| See active users right now |
| Most viewed pages |
| Traffic by source/medium |
| Summary with optional period-over-period comparison |
Tool | Description |
| List all Search Console sites you have access to |
| Flexible search analytics query |
| Top search queries by clicks |
| Top pages in search results |
| Check indexing status of any URL |
| List submitted sitemaps |
Tool | Description |
| List accessible ad accounts |
| Account-level spend/performance |
| Campaign metrics |
| Ad set metrics |
| Individual ad metrics |
Tool | Description |
| Execute raw GAQL queries |
| Campaign performance metrics |
| Keyword metrics + quality score |
| Ad group metrics |
| Account totals + top campaigns |
Related MCP server: google-search-console
git clone https://github.com/YOUR_USERNAME/mcp-marketing-analytics.git
cd mcp-marketing-analytics
npm install
npm run buildRun the interactive setup wizard:
npm run setupOr manually create a .env file (see .env.example).
Go to Google Cloud Console
Create a project and enable:
Google Analytics Data API
Google Analytics Admin API
Google Search Console API
Google Ads API (optional)
Create OAuth 2.0 Client ID (Web application type)
Add http://localhost:3847/callback as an Authorized redirect URI
Run npm run setup to complete the OAuth flow and get your refresh token
Go to Meta Graph API Explorer
Generate a token with ads_read permission
Exchange for a long-lived token (60-day expiry)
Create a Manager Account
Get a Developer Token from Tools & Settings → API Center
Note your Customer Account ID (10 digits)
Edit %APPDATA%\Claude\claude_desktop_config.json (Windows) or ~/Library/Application Support/Claude/claude_desktop_config.json (Mac):
{
"mcpServers": {
"marketing-analytics": {
"command": "node",
"args": ["/absolute/path/to/mcp-marketing-analytics/dist/index.js"],
"env": {
"GOOGLE_CLIENT_ID": "your-client-id",
"GOOGLE_CLIENT_SECRET": "your-client-secret",
"GOOGLE_REFRESH_TOKEN": "your-refresh-token"
}
}
}
}claude mcp add marketing-analytics -- node "/absolute/path/to/mcp-marketing-analytics/dist/index.js""List my GA4 properties"
"Show me top pages for the last 30 days"
"What are my top search queries on Google?"
"How are my Meta ad campaigns performing?"
"Compare this month's traffic to last month"
All API credentials are passed via environment variables. Only configure the APIs you need — unconfigured APIs will show a helpful message instead of crashing.
Variable | Required | Description |
| For Google APIs | OAuth 2.0 Client ID |
| For Google APIs | OAuth 2.0 Client Secret |
| For Google APIs | OAuth 2.0 Refresh Token |
| No | Default GA4 property (can be set per-request) |
| No | Default GSC site (can be set per-request) |
| For Meta Ads | Long-lived access token |
| No | For app secret proof |
| No | For app secret proof |
| For Google Ads | From API Center |
| For Google Ads | 10-digit account ID |
| No | MCC manager account ID |
You don't need to hardcode property IDs or site URLs. Use the discovery tools:
ga4_list_accounts → shows all GA4 properties you can access
gsc_list_sites → shows all Search Console sites you can access
Pass propertyId or siteUrl to any tool to query a specific account
MIT
List all GA4 properties accessible to the authenticated Google account. Use this to discover Property IDs for use with other ga4_* tools.
Inputs: no named parameters.
Get a realtime GA4 report showing current active users and activity from the last 30 minutes.
Inputs: limit, metrics, dimensions, propertyId.
Run a custom GA4 report with specified dimensions, metrics, and date range. Use GA4 API names like 'activeUsers', 'sessions', 'pagePath', 'sessionSource'. Dates: YYYY-MM-DD or relative like '7daysAgo', '30daysAgo', 'today', 'yesterday'.
Inputs: metrics (required), dimensions (required), limit, endDate, +5 more.
Get the top pages by views for the specified date range. Returns page path, title, views, users, and average session duration.
Inputs: limit, endDate, startDate, propertyId.
Get traffic source breakdown showing sessions, users, and engagement by source/medium for the specified date range.
Inputs: limit, endDate, startDate, propertyId.
Get a summary of key user metrics: total users, new users, sessions, page views, bounce rate, avg session duration, and engagement rate. Optionally compare with the previous period.
Inputs: endDate, startDate, propertyId, comparePreviousPeriod.
Get a high-level summary of the Google Ads account: total spend, clicks, impressions, conversions, and key averages, plus top 5 campaigns by spend.
Inputs: dateRange.
Get performance metrics for Google Ads ad groups: impressions, clicks, cost, conversions grouped by ad group within campaigns.
Inputs: limit, orderBy, dateRange, campaignName.
Get performance metrics for Google Ads campaigns: impressions, clicks, cost, conversions, CTR, CPC, and conversion rate.
Inputs: limit, status, orderBy, dateRange, +1 more.
Get performance metrics for Google Ads keywords: search term, match type, clicks, impressions, cost, conversions, and quality score.
Inputs: limit, orderBy, dateRange, campaignName.
Execute a raw Google Ads Query Language (GAQL) query. Use for custom queries not covered by other tools. Example: SELECT campaign.name, metrics.clicks FROM campaign WHERE segments.date DURING LAST_30_DAYS
Inputs: query (required).
Inspect a URL to see its Google indexing status, crawl info, mobile usability, and any issues. Useful for debugging why a page isn't appearing in search.
Inputs: url (required), siteUrl.
List all sitemaps submitted for the site, including their status, last download date, and number of URLs submitted/indexed.
Inputs: siteUrl.
List all Search Console sites/properties accessible to the authenticated Google account. Use this to discover Site URLs for use with other gsc_* tools.
Inputs: no named parameters.
Query Google Search Console search analytics data with flexible dimensions and filters. Returns clicks, impressions, CTR, and average position.
Inputs: endDate (required), startDate (required), siteUrl, rowLimit, +5 more.
Get the top pages appearing in Google Search results, sorted by clicks. Shows page URL, clicks, impressions, CTR, and average position.
Inputs: endDate (required), startDate (required), limit, siteUrl, +1 more.
Get the top search queries driving traffic to your site, sorted by clicks. Shows query, clicks, impressions, CTR, and average position.
Inputs: endDate (required), startDate (required), limit, siteUrl, +1 more.
Get a high-level performance overview for a Meta ad account: total spend, impressions, clicks, CTR, CPC, CPM, and conversions for the specified date range. Optionally break down by age, gender, country, placement, or device.
Inputs: endDate (required), accountId (required), startDate (required), breakdown.
Get performance insights for individual ads. Shows each ad's name, ad set, campaign, spend, impressions, clicks, and conversions.
Inputs: endDate (required), accountId (required), startDate (required), limit, +2 more.
Get performance insights for ad sets within a campaign or account. Shows spend, impressions, clicks, and conversions per ad set.
Inputs: endDate (required), accountId (required), startDate (required), limit, +1 more.
Get performance insights for all campaigns in a Meta ad account. Shows each campaign's spend, impressions, clicks, CTR, CPC, and conversions.
Inputs: endDate (required), accountId (required), startDate (required), limit, +1 more.
List all Meta/Facebook ad accounts accessible with the current access token. Returns account ID, name, status, currency, and timezone.
Inputs: limit.
Get step-by-step instructions for configuring the Marketing Analytics MCP server. Shows which APIs are configured and provides detailed setup guides for any that are missing.
Inputs: no named parameters.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 23 tools
Tools are clearly organized by platform prefixes (ga4_, gsc_, meta_, gads_), and each tool targets a distinct resource or action. There is minor overlap within platforms (e.g., ga4_top_pages vs. ga4_run_report, gsc_top_pages vs. gsc_search_analytics), but descriptions clarify when to use each.
Names follow a strong platform-prefix convention (ga4_, gsc_, meta_, gads_) which aids navigation. Within each platform, naming is mostly descriptive noun phrases (e.g., ga4_top_pages, meta_campaign_insights) with some verbs (list, run, inspect), creating minor inconsistency but no chaos.
23 tools is on the heavy side, but appropriate for a multi-platform analytics server covering four distinct services (GA4, Search Console, Meta Ads, Google Ads). Each platform receives a focused set of 5-6 tools, making the count reasonable for the scope.
The tool set provides solid read-only coverage across all four platforms: account discovery, standard reports, custom queries, and granular insights. Some advanced features (e.g., GA4 event diagnostics, Meta ad creative details) are absent, but core analytics workflows are well supported.
Read-only Google Ads, GA4 and Search Console data for Claude, with no Google Cloud setup.
GA4, Google Ads and Search Console in Claude. Read-only OAuth, multi-account for agencies.
Live Google Ads, GA4, Search Console, Meta Ads and GBP data in Claude, ChatGPT and any MCP client
Ask Claude about your ads: Meta, Google, TikTok, LinkedIn, GA4 & Shopify. No AI credits.
