Why learn all these skills? Matthew Roche, Power BI CAT team and governance superstar, has created a principle to understand where to transform data. Pushing upstream may reduce the complexity of your DAX.
Learn what this means and why it is so important.
Source
The source of this article was a post in LinkedIn:
Excellent, and that’s what I’ve always been saying.
Everyone thinks about learning Power BI (or for example a programming language) by watching a video on Youtube, and doing exactly the same*.
They forget the most important, at least the vast majority - the importance of data modeling!
It is always where there is the most resistance, and some even think that it is not necessary to study.
They end up, sooner or later, realizing that they should have done this, as they cannot make logical connections or correlations between the data, for example.
Anyway, congratulations for raising this issue, it can help a lot those who are starting.
* Note: I am not against learning from examples - for me, is the best way to get better, and start. However, ALSO learning the fundamentals is essential.
The Power Query M language, commonly referred to as “M,” is a powerful data transformation and data preparation language designed specifically for Microsoft’s Power Query, an ETL (Extract, Transform, Load) tool integrated into the Power BI, Excel, and other Microsoft products.
M was developed with the primary purpose of enabling users to clean, reshape, and combine data from various sources, facilitating the process of transforming raw data into meaningful insights. By leveraging the power of M, users can create custom functions, conditional columns, and advanced data transformations that go beyond the capabilities of traditional spreadsheet applications.
STAR Schema
Data modelling
Power Query
Roche’s Maxim of Data Transformation
Source