Securing attributes

In this section, we will take a wholly different look at security in Power BI models. In the previous sections, we focused on ways to restrict the visibility of rows in tables of the model. This is the most common security need but, sometimes, other forms of security are needed. If you think of row-level security as "horizontal" security, looking at your data, then it makes sense to consider the possibility of "vertical" security. In other words, can we secure columns, or attributes?

The case for secured attributes

Securing a Power BI model with RLS is only really needed when the model is used by a broader audience. If you have a model used only by C-level executives, you probably won't need to have RLS at all, as each user is allowed to see all the data. Only when the audience grows larger does the need arise to segment the data based on geography, customer segment, or organizational structure, as discussed in the previous sections...