GoodData, a leader in AI analytics and decision intelligence, announced the public launch of its MCP Server. As organisations adopt AI in analytics, most tools remain limited to query generation, leaving teams to manually manage metrics, dashboards and business logic. MCP Server moves AI beyond analysis, enabling governed, end-to-end analytics execution and delivering 10–50x faster time to value.
The MCP Server is designed for AI developers, and BI and data teams who want to build and manage analytics faster with AI. It allows AI to build and operate analytics in the same way a skilled human team would, but faster and without operational bottlenecks.
Using the Model Context Protocol (MCP), AI agents and large language models (LLMs) can connect directly to GoodData and execute analytics across the full lifecycle. They can work with governed analytics assets, including semantic models, metrics, dashboards and alerts, without relying on screenshots, SQL copy-and-paste or fragile UI workflows. In practice, this means AI can build, update and run analytics processes and agentic workflows automatically, while respecting the same rules and controls as human users.
From AI-assisted analysis to analytics execution
GoodData’s MCP Server shifts AI from interacting with analytics to executing within it. Rather than layering AI on top of dashboards or query interfaces, MCP exposes analytics as executable infrastructure.
By combining analytics-as-code, governed APIs and LLM-based coding, MCP Server allows AI to create, modify and validate analytics assets directly. Definitions remain consistent as they evolve, analysis runs continuously and changes propagate safely, without requiring manual intervention at every step.
All execution takes place under the same security, permissions and governance model used by human teams. Business rules are enforced by the system rather than relying on individual knowledge, reducing operational risk while increasing speed and reliability.
How teams use MCP Server
With MCP Server, analytics and BI assets become programmatic resources that AI can work with directly:
- Accelerate BI development with AI: Analytics-as-code allows AI to build and update analytics automatically, reducing BI backlogs and eliminating manual UI-driven work.
- Enable continuous AI-driven analysis: Once analytics are defined, execution is ongoing, data is queried, results are computed, dashboards are updated, alerts are scheduled and logic remains in sync.
- Extend analytics to any AI agent: Any MCP-compatible agent can safely use GoodData’s full analytics capabilities, modelling, metrics, queries, alerts and validations, under the same governance controls as human experts.
Why this works now
This shift is possible because three advances have converged: MCP provides a standard execution layer for AI, analytics-as-code makes analytics programmable, and modern LLMs can reliably operate complex systems within defined constraints. Together, they transform analytics execution from a linear, people-bound process into a scalable platform capability.
Source: GoodData
