Unity Catalog 0.6.0 introduced metric views as governed catalog objects for semantic layers.

In August 2026, the open source Unity Catalog 0.6.0 release formally introduced metric views and SQL views as governed catalog objects, providing native catalog level infrastructure for semantic layers alongside Apache Spark 4.2 support [5]. Databricks also expanded migration pathways to accelerate adoption, introducing automated capabilities within Genie to convert existing PowerBI dashboards directly into Databricks metric views [6].

Community discussions examined how metric views reshape traditional medallion architectures. Practitioners argued that gold tables should no longer serve as the terminal point of data pipelines, because centralized metric views prevent different teams from calculating the same business metrics twice in downstream dashboards [4, 11]. Technical analyses explored the practical evolution of metric definitions, outlining workflows that combine YAML configuration files, visual user interfaces, Genie code generation, and table materialization techniques [10].

Teams also prepared their data environments for conversational analytics and external consumption. Community members shared guidance on data modeling and ontology design required to ground text to SQL agents like Genie in structured catalog data [3, 7]. At the same time, developers raised questions about accessibility, exploring how to query Unity Catalog metric views directly from external custom applications operating outside Databricks [8], while tracking roadmap developments for spreadsheet integrations including an Excel add-in [2].