Power BI Weekly

Issue 366

14th July 2026

Welcome to the 366th edition of Power BI Weekly! No official announcements from the Power BI team this week, but on the Fabric updates blog, we've been told about the VNET Data Gateway Autoscaling (now in Preview) which dynamically adjusts active nodes based on workload demand to balance performance and costs within administrator-defined limits. Now onto the community content:

Kurt Buhler has written about how to write good descriptions for semantic model columns and measures, arguing descriptions are now first-class metadata rather than optional hygiene. In particular, Kurt makes the point that a description written for the benefit of Copilot is often a bad description for a human hovering over a measure, so write each one with a specific audience in mind.

Moving on, there are a few of these this week, but Reza Rad has a video demoing no-code Power BI report authoring using GitHub Copilot, the Fabric Skills Report Authoring skill, and the Power BI Desktop Bridge, building a full report from natural language prompts alone (benefiting from the new Fabric Skills and Power BI Desktop Bridge tooling. Both of which are relatively hot-off-the-press, but definitely worth experimenting with.

My colleague James Broome has laid out workspace topology patterns for Microsoft Fabric, from a single workspace through to full medallion-layer splits (including separating Reporting workspaces from Data Engineering workspaces), along with the tipping points that push you from one to the next. And on the DAX/reporting side, Marco Russo has published a piece on dynamic formatting by hierarchy level using ISINSCOPE and ISATLEVEL, applying different colour logic to a matrix visual by year, quarter, or month.

Finally, Chris Webb has written up the benefits of using Direct Lake mode in Power BI: faster, cheaper refreshes, no more memory or timeout limits, and reuse of the same OneLake data across models and engines. But there are plenty of caveats that are noted in both Chris' and others' related blogs. Also, take a read of the comments section - some interesting points that complement the core blog content.

Data Prep

  • Developing the Semantic Model For Microsoft Fabric App You'll learn to develop a comprehensive semantic model for Microsoft Fabric apps, including data ingestion from OneDrive via Dataflow Gen2 and Lakehouse, and key quirks of Direct Lake to avoid troubleshooting.

Data Modeling

  • Writing good descriptions for semantic model columns and measures Writing clear descriptions for semantic model columns and measures improves usability for developers, users, and AI by explicitly revealing hidden information and customizing content for each audience while treating descriptions as vital first-class metadata.
  • Optimising DAX: Model Design Comparisons and Testing Tips The article finds the star schema best for compression and performance, flat tables surprisingly efficient with run-length encoding, and the header/detail model least optimal due to high-cardinality links.
  • Power BI + Claude: Análisis de Estado de Resultados con IA (DAX desde cero) [Spanish] This tutorial shows how to integrate Anthropic’s Claude AI with Power BI via the Model Context Protocol for automated data updates, DAX measure creation, and dynamic filtering of financial statements without manual input.
  • The Power BI Expert Is GONE. Now What? This video shows how Copilot for Power BI speeds up understanding and maintaining semantic models by clarifying existing logic, documenting measures, and verifying capabilities without requiring the original developer.
  • DAX vs Excel Formulas: What’s the Real Difference? DAX and Excel formulas differ fundamentally-Excel relies on cell references and manual updates, while DAX operates within a data model using measures and dynamic contexts, making Power BI reports more scalable and responsive than spreadsheet-based solutions.

Report Authoring and Interactivity

Deployment, Security and Operations

  • VNET Data Gateway Autoscaling (Preview) VNET Data Gateway autoscaling dynamically adjusts active nodes based on workload demand to balance performance and costs within administrator-defined limits.
  • The Benefits Of Using Direct Lake Mode In Power BI Direct Lake Mode in Power BI provides faster, cheaper refreshes and reduces development effort by using Spark, but query performance and cloud usage costs depend on data volume and refresh frequency.
  • Microsoft Fabric Workspace Topology Patterns Understanding workspace topology patterns in Microsoft Fabric is crucial for setting up secure, scalable, and manageable architectures from the start, avoiding costly refactors later.

General

Get Power BI Weekly in your inbox every Tuesday

Sign up to receive the email every week and keep on top of all the announcements.

Don't forget to confirm your subscription. We have a double opt-in sign up process to prevent spam.

We will never provide your details to any third parties. We don't spam.

Power BI Weekly Logo