In the PowerPivot book and in the PowerPivot Workshop I mentioned the difference between implicit and explicit measures. The use of implicit measures might be confusing because at the end they are automatically created and maintained by the user interface of PowerPivot, but in reality the “real engine” doesn’t know the difference between them.
Cathy Dumas, which is a member of the Analysis Services team and has been working on BISM Tabular and PowerPivot development has just published a post about the import of a PowerPivot workbook into a BISM Tabular model. I suggest you the reading of her previous posts as well – subscribing to her blog is a very good idea, she already published an interesting discussion about the use of MSBuild to automate deployment of a Tabular project and other blog posts deserve a read.
The interesting part for my current post is that in BISM Tabular you only have explicit measures. You have to explicitly create the measures you want to expose and the import operation automatically converts implicit measures into explicit ones. Of course, only those columns that have been used in a PivotTable as measures become explicit measures in the BISM Tabular project and you have to manually create other measures if you think your users will be interested in using them. In fact, when you browse a BISM Tabular model in Excel, you use the regular PivotTable user interface and you don’t use the PowerPivot Field List, that allows you to create new implicit measures “on the fly” and to move attributes in slicers in a very easy way. I hope this will be improved in the future, but in the meantime you have to model your data figuring out that your data model will be shown as a dimensional model and not like a “tabular” one!