Customizable tag/label for Report/Semantic model to highlight untrusted data Source

Today with self-service Capabilities user are able to create composite models on top of trusted Semantic models were they can merge with untrusted data source such as excel.

It is very difficult to challenge users not to use excel as data source and often create frictions. So I suggest to have a customizable label or Tag in addition to certification where you can have a warning message when a

A user open Report/Semantic model which warns him that some of the data source are not trusted .