How Scottish Water Made Its Capital Investment Data Conversational With Databricks Genie

Scottish Water’s Capital Investment teams previously relied on reports or specialists to access project status, financials, and risks, creating delays and duplicated work. Databricks Genie now allows users to ask questions in Microsoft Teams and receive immediate, governed answers without navigating complex reports. The shift addresses an access problem rather than a data shortage, streamlining workflows for analysts and delivery teams alike. SPARK, Scottish Water’s branded interface, integrates Genie directly into Teams via Copilot and a supervisor agent, ensuring answers are derived from governed data in Unity Catalog.

The implementation prioritised trust and consistency by grounding Genie in governed Unity Catalog data and shared semantic definitions aligned with business logic. Scottish Water configured the Genie space around its own terminology and real user questions, avoiding generic setups to ensure reliability. Monitoring was built in to track adoption, answer quality, and performance, while the solution was designed for safe, consistent promotion across environments. This approach ensures the tool remains reliable and scalable rather than a one-off experiment.

For leaders, the tool reduces reporting friction by providing faster visibility into project data without manual extraction. Delivery teams and data users gain direct access to governed insights within their existing workflows, eliminating the need to shift between tools. The natural-language interface replaces time spent searching for reports with immediate, actionable answers grounded in trusted data. Scottish Water’s approach demonstrates how governed data combined with Genie can democratise access to business-critical information.

The result is a practical tool that transforms how teams interact with project data. By embedding Genie into Microsoft Teams via SPARK, Scottish Water eliminated barriers between users and the information they need. The solution’s governance and monitoring ensure answers remain consistent and explainable, while its design supports safe scaling across the organisation. This implementation highlights the potential of conversational analytics when paired with robust data governance.