Microsoft Fabric adds Materialized Lake Views to store query results as physical tables
Reviewed by Helen Jones
Microsoft Fabric now supports Materialized Lake Views. These store the output of SQL queries as physical Delta tables inside OneLake. The new REFRESH MATERIALIZED LAKE VIEW command updates those tables. Fabric picks between no-refresh, incremental, or full refresh based on detected changes. The feature works best with declarative SQL transformations. Procedural logic still needs Spark notebooks.
Before this, analysts who needed fast, repeatable results either rebuilt joins and aggregates every time in Power Query or kept static Excel extracts on SharePoint. Both approaches created hidden refresh chains and version drift. Now the same logic can live as a governed table inside OneLake. The table updates on its own schedule, carries lineage, and can feed Power BI semantic models directly. The cost is that refresh policy and data-quality rules are now set at the lake layer, not inside the report.
Analysis
Map every nightly Power Query merge or aggregation step that feeds your dashboards and move the stable ones to Materialized Lake Views before automatic policies lock in refresh behaviour you cannot explain.
Pulse published by Collab365 Spaces, reviewed by Helen Jones on . Cite as "Microsoft Fabric adds Materialized Lake Views to store query results as physical tables", Collab365 Spaces. 1 source referenced.