If getting data from a relational source such as SQL Server, use views to abstract the report from the physical layout of database tables and never write queries inside reports; from a maintenance and development standpoint, it is simpler and faster to update a view than it is to edit a query inside a report and then republish the report.

Also, if available, create a separate schema for the views for grouping and security purposes.  This is the recommended architecture:

Click here for a full set of Power BI best practices.

#PowerBI Best Practice: use dedicated views and schema

Powered By the Tweet This Plugin

Tweet This

Please follow and like us:

There are three ways to connect to data: Import, Live Connection, and DirectQuery (note: currently in preview are hybrid methods).  The choice of method will depend on various considerations, such as the size of the data, the number of users, and the data refresh needs.  Make sure an appropriate assessment is made before the report is built, as switching from one method to another is not always possible or straightforward.

The following chart summarizes the options on several dimensions.

*This is for Power BI Pro.  Power BI Premium has more flexible refresh and size limits.

Click here for a full set of Power BI best practices.

#PowerBI Best Practice: carefully choose the data source connection method

Powered By the Tweet This Plugin

Tweet This

Please follow and like us: