I have recently read this book (published in 2017) by Ferrari and Russo (aka "The Italians") of sqlbi.com and found it extremely informative with a very practical focus. To analyze data that resides in, or comes from more than one table, you will frequently need to model the data regardless of the technology platform you are using. This book describes how to model data for a variety of analytical scenarios. From a generic data modeling viewpoint, the book appears to incorporate aspects of the Kimball Techniques but does so in the context of the workings of the Excel/Power BI Data Model. A common thread is the focus on creating your model so that the use of complex DAX calculations is substantially minimized.
This is not a review of the book and I have no vested interest in promoting it. But if you are working with the Excel/Power BI Data Model and unsure how best to build your model to meet your analysis needs, you might find this book both useful and readable.








