Using Tabular Models - SQL Server Analysis Services 2012 Cube Development - page 314

Modeling the data
Now that the data has been loaded, there are a few modeling techniques that should
be applied to the model. Due to the mash up nature of the Tabular Models, you will
likely need to do the following common clean up operations:
• Update column names
• Fix data types and formats
• Add relationships
You will learn how to do each of these in this recipe.
Getting ready
Before you can model data, you will need data in your model. If you have not loaded
data yet, refer to the previous recipe on loading data.
How to do it…
You will be working in SSDT to make the changes.
1. Our first change is to update the column names. The column names should
be user-friendly and easy to understand.
2. You can rename the column by right-clicking on the column and selecting Re-
name Column from the shortcut menu. In some complex scenarios, you may
be required to rename all of the columns. In particular, this will likely be re-
quired for text loads that may be loaded with column names; for example,
Column1, Column2, and so on.
Note
As a best practice, you should update column names when the data is
loaded and before you begin updating the model. This will prevent issues
with calculations that are added to the model later and are dependent on
the column names.