If you are a BI consultant and data modeler, for sure you have been in a situation in which you have created a complex data model ignoring the best practice advice of BI experts. Your model combines financial, sales and inventory data whereby incorporating these into a complex, but nicely working data model. However your model faces one important issue – its complexity and number of measures and tables are frightening for end users. When business analysts connect to your model, they see all tables, columns and measures, and KPIs, and not just ones they work with. Often this causes confusion.
When working with SSAS, the solution to the problem is to use perspectives. The perspectives split our model into subsets based on different criteria, a functional area for instance. We can define perspectives to be used by the sales department, the financial department or others and thus make the usage of the model easier.
But what is the solution when the semantic model is created in Power BI?
The August version of Power BI made it possible to use perspectives in Power BI without the SSAS model. The steps of creating perspectives are simple and clear:
- perspectives are created in Power BI Desktop using Tabular Editor as an external tool;
- In the Tabular Editor you define which tables, columns and measures are visible in the different partitions;
- For each Power BI page you define the usage of a perspective (subset of meta data);
- The Power BI report is published to the service.
- As a result of the above actions the visual personalization functionality can be used more intuitively as user can only see those data which the Power BI author has defined as suitable and meaningful in the page context.
More information about Power BI and the Advanced Analytics solutions of FTS can be found here here.