Case study · Logistics & Mobility
AI-native analytics: grounding LLM agents in governed metrics with MCP
Connecting LLM agents to a governed metric store through the Model Context Protocol so teams can self-serve analytics conversationally.
What
An AI-native analytics layer that connects LLM agents to Porter's governed Metric Store through the Model Context Protocol (MCP), so business users can ask questions in conversation and get governed answers.
Why
Self-serve dashboards answer the questions someone predicted. Business users also ask questions nobody predicted, and those still become report requests. LLM agents can answer them, but only if the answers are grounded in the same definitions as the rest of the company.
How
- Governed metrics first. Definitions come from the Metric Store and Semantic Reporting Layer that already power self-serve reporting (see the Metric Store case study).
- MCP as the interface. MCP exposes governed metrics to agents through a standard, auditable interface, not raw tables.
- Guardrails inherited. Access controls, PII masking and audit trails from the governance program apply to agent queries too.
- Capability in-house. The Analytics Engineer function was set up partly to build and maintain this capability.
Who
Senior Analytics Manager, with the Analytics Engineer function and Data Engineering.
When
In progress at Porter.
Result
Governed metrics through conversation: answers stay consistent with dashboards because both read the same definitions. This program is in progress, and I will publish measured results as they land.
An LLM agent is only as trustworthy as the metrics it is grounded in.