Direct Lake is a great addition to the established Power BI storage modes, but simply using it like the others will not serve you well.

In this benchmark-driven session, I will share what I learned from rigorous Direct Lake testing in Microsoft Fabric. I will show how the underlying data design affects performance, how to keep frequently used data available quickly, how much it can cost overall, and which patterns Direct Lake follows when serving data to Power BI reports.

You will leave knowing when Direct Lake makes sense, how to design for it, and how to avoid paying for performance you do not actually get.