data-slots=image, heading, text, buttons

data-background=rgb(37, 37, 37)

Analytics MCP servers

MCP servers for Adobe Analytics, Customer Journey Analytics and Marketing Campaign Analytics (formerly known as Mix Modeler)

Get started

data-slots=heading, links

Resources

Overview

This documentation covers the Adobe-hosted MCP servers for Adobe Analytics, Customer Journey Analytics, and Marketing Campaign Analytics. These servers connect MCP-compatible clients (like Claude, ChatGPT, or Cursor) directly to your analytics environment. Once connected, your client can invoke product-specific tools to query data, explore trends, build segments, and manage components using natural language as part of an LLM or agent-driven workflow.

The MCP servers share a standardized connection model with consistent endpoints and a common onboarding flow. See Getting started to learn how to connect the desired MCP server to your agent of choice, or the Adobe Analytics or Customer Journey Analytics tool references to explore each server's capabilities.

data-slots=text

data-variant=info

Currently, the Marketing Campaign Analytics MCP server is only available through Adobe CX Enterprise Coworker.

Additional Information

If you are looking for analytics data collection strategies for your own MCP servers, see Collect analytics and apply personalization in MCP clients in the Adobe Experience Platform Data Collection documentation.

This user guide adheres to Adobe's Code of Conduct. Contributions are encouraged and appreciated. See Adobe's Code of Conduct and Contribution Guidelines on GitHub for more information.

data-variant=warning

data-slots=text, text

The Model Context Protocol (MCP) is an open-source standard that can introduce security and reliability risks. Connecting MCP clients or servers to Adobe products is a customer-elected configuration. Customers are responsible for evaluating the security and suitability of any MCP integration. Adobe is not responsible for issues arising from misconfiguration, misuse, third-party vulnerabilities, or unintended actions performed through MCP-enabled workflows.

To reduce risk, test integrations in a sandbox environment before production use. Validate all MCP-initiated actions and responses before acting on them.

This user guide adheres to Adobe's Code of Conduct. Contributions are encouraged and appreciated. See Adobe's Code of Conduct and Contribution Guidelines on GitHub for more information.