MCP server

View how to connect an AI assistant to Drivetrain, ask about your own metrics and compare them with industry benchmarks.

How the MCP server works in Drivetrain

Everything builds on connecting your AI assistant to Drivetrain. From there, ask about your metrics, compare them with benchmarks, find ways to improve them and pass the results on to other tools.

1
Connect an AI assistant to Drivetrain
- Connect any AI assistant that supports the Model Context Protocol (MCP), such as the Claude desktop app, to your Drivetrain data.
- Once connected, the assistant can draw on your financial data in Drivetrain when it answers a question.
2
Ask questions about your own metrics
- Ask a business question in plain language, for example "What's my revenue per FTE?"
- The assistant checks which metrics exist in Drivetrain and uses the one that matches the question.
- If no matching metric exists, it pulls the underlying data, for example revenue and headcount, and calculates it.
- Get values month by month for the period you ask about.
3
Compare your metrics with industry benchmarks
- Ask how a metric compares with companies in your revenue range.
- The assistant searches the web for current industry benchmarks and lays them out in a table next to your own number.
- See whether you sit above, within or below each peer group.
4
Get ideas to improve a metric
- Ask how to improve a metric, and the assistant suggests a plan based on best practices, benchmarks and your current data.
- Ideas are grouped by priority, from quick wins to larger projects, along with what to delegate and what to avoid.
5
Combine Drivetrain with other MCP servers
- Connect other MCP servers to the same AI assistant to act on the results.
- For example, draft an email to your leadership team or board with the Gmail MCP, or update a shared Notion page with the research.

How Drivetrain simplifies working with AI assistants

Ask the AI assistant your team already uses about your Drivetrain numbers, without exporting files first.
Get answers built on the metrics you already defined in Drivetrain, with the metric behind each answer shown.
Compare your own numbers with industry benchmarks in the same conversation.

FAQs

Which AI assistants can connect to Drivetrain's MCP server?

Any AI assistant that supports the Model Context Protocol (MCP), such as the Claude desktop app. Once it is connected, the assistant can draw on your financial data in Drivetrain.

Does Drivetrain's MCP server use the metrics we already built in Drivetrain?

Yes. When a matching metric exists in Drivetrain, the assistant uses it. If it doesn't, the assistant pulls the underlying data, such as revenue and headcount, and calculates the metric.

Can Drivetrain's MCP server compare our numbers with industry benchmarks?

Yes. The assistant searches the web for current benchmarks for companies of your size and revenue range, and lays them out in a table next to your own number. The benchmark data comes from the assistant's web search, not from Drivetrain.

Can Drivetrain's MCP server help us improve a metric, not just report it?

Yes. Ask how to improve a metric and the assistant suggests a plan based on best practices, benchmarks and your current data, grouped from quick wins to larger projects.

Can Drivetrain's MCP server work alongside other MCP servers?

Yes. You can connect other MCP servers to the same assistant, for example Gmail to draft an email to your leadership team or board, or Notion to update a shared page with the research.