> For the complete documentation index, see [llms.txt](https://integrations.drivetrain.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://integrations.drivetrain.ai/extending-drivetrain/using-the-drivetrain-mcp-server.md). # Using the Drivetrain MCP Server ## Using the Drivetrain MCP Server The **Drivetrain MCP server** allows you to connect Drivetrain to **MCP-compatible clients** such as ChatGPT and Claude. Once connected, Drivetrain metrics are exposed as **callable tools**, allowing MCP-compatible clients to retrieve metric values across time, dimensions, and versions. This provides a new way to work with your Drivetrain metrics: in addition to the application interface, you can now access the same trusted metrics through natural-language prompts in an external client. #### What is an MCP server? An **MCP (Model Context Protocol) server** is a service that exposes tools - functions that an AI client can call to fetch or process information. When connected, the client determines which tool is relevant to a prompt, calls it, and returns the result in a structured format. For Drivetrain, this means: * **Metrics, dimensions,** and **versions** are available as callable tools. * Clients such as [Claude Desktop](https://claude.ai/download) can **fetch Drivetrain metrics**, perform derived calculations, and present results as text or charts. * Metrics can be combined with outputs from other MCP servers to power **cross-source analysis**. {% embed url="" %} #### Scope of the Drivetrain MCP Server The **Drivetrain MCP server** provides access to: * Metric values * Across time * Filtered by dimensions * Scoped by versions * Including derived metric calculations It does **not** provide access to: * Raw datasets * Board editing * Model editing * Schema changes * Dataset-level querying #### Example workflow Once connected, you can use natural-language queries to work with Drivetrain metrics. For example: Prompt: `What’s my Revenue per FTE?` Here’s what happens in the background: * The system first checks whether a **“Revenue per FTE”** metric exists in Drivetrain. * If it exists, the metric is fetched directly. * If not, the system fetches the **Revenue** and **FTE** metrics separately and computes the derived value (Revenue ÷ FTE). * Actual values for the requested time period are then retrieved and displayed—for example, Revenue per FTE across 2025, showing month-to-month trends. This workflow can be extended to cover deeper analysis: * **Benchmarking:** You can also prompt: `How does this compare to industry benchmarks for companies of my revenue range?` * The system automatically searches and consolidates **up-to-date benchmark datasets**. * Results are filtered to match your **company size** and **revenue tier**. * The output is a **structured comparison table**—for instance, showing that $168K revenue per FTE is above the private SaaS median of $129K but below the $200K–$230K range for larger SaaS companies. * **Strategic recommendations:** You can follow up with the prompt: `How can I improve my revenue per FTE?` * **Best practices** and **strategies** are curated from trusted sources and contextualized with your metrics. * Recommendations may include **quick wins** (e.g., pricing optimization, automation, expanding revenue from existing customers) and **longer-term initiatives** (e.g., employee development, performance management, strategic hiring). * A **roadmap** is outlined with progressive targets and timelines. * **Extending with other MCP servers:** By connecting additional servers, you can automate downstream workflows. For example * Use a **Gmail MCP** server to draft and send an update email to your leadership team. * Use a **Notion MCP** server to automatically update a shared page with benchmark comparisons and strategies. In this way, a **simple question evolves into a complete workflow**—from retrieving Drivetrain metrics, to comparing them with external benchmarks, to generating strategic recommendations and sharing them with stakeholders. #### Why use the MCP server? The **Drivetrain MCP server** allows you to: * **Query metric values** using **natural-language prompts** * **Retrieve metrics** across time, dimensions, and versions * **Compute** derived metrics * Use Drivetrain metrics inside **external AI-driven workflows** In the next section, you’ll learn how to connect common MCP clients—such as ChatGPT and Claude—to the Drivetrain MCP server. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://integrations.drivetrain.ai/extending-drivetrain/using-the-drivetrain-mcp-server.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.