Currently, when exporting data from a matrix, the export counts the underlying data rows, not the rows displayed in the matrix view. While this works in some contexts, it presents challenges when working with large datasets. The ability to export the matrix view rows rather than the underlying data rows would be a significant improvement, particularly when managing large and complex datasets. I have invested a considerable amount of time in developing a comprehensive architecture for row-level data reconciliation. As you're likely aware, when data is unpivoted, the difference between exporting matrix rows versus underlying data rows can be substantial—potentially expanding from 400,000 rows to 5 million rows, depending on the number of columns. For data professionals and business power users, the data warehouse serves its' purpose, but the reporting solution in Power BI is crucial for bridging the gap between IT and business stakeholders. This change would not only enhance our current data migration and reconciliation efforts, but also provide users with a more streamlined and efficient experience when leveraging Power BI’s powerful visualization capabilities. I believe this enhancement is especially relevant now with the advancements in Power BI and the integration of Fabric for staging physical databases. It seems feasible to adopt a semantic model that accommodates larger outputs and allows for a customized switch between matrix view rows and underlying data row counts. This improvement would further elevate Power BI’s capabilities, and ensure that users can work with data in a way that better matches their needs and workflow. It would also help address challenges that are currently making solutions like Tableau 👎 more appealing for large-scale reporting tasks. Thank you for considering this suggestion. I look forward to hearing your thoughts on this potential enhancement. Best regards,
... View more