Working with Attributes

Attributes describe characteristics of data, such as the size and color of products. You can use attributes to group and analyze members of dimensions based on their characteristics. For example, you can analyze product profitability based on size or packaging, and you can make more effective conclusions by incorporating market attributes, such as the population size of each market region, into your analysis.

You can assign attributes to sparse dimensions only. You can't assign attributes to label-only members. Attribute dimensions don't have aggregation properties because parents are dynamically calculated.

The Account dimension is usually defined as dense, so you can't assign attributes to it unless it's changed to sparse for all cubes. If you change a dimension from sparse to dense, all attributes and attribute values for that dimension are automatically deleted.

Attributes can have data types of text, date, Boolean, and numeric, as described in Understanding Attribute Data Types.

Attribute names must conform to guidelines listed in Naming Restrictions.

To create or edit attributes in the Simplified dimension editor:

1. From the Home page, click Application, click Overview, and then click Dimensions.
2. The Cube drop-down list defaults to All. Click the down arrow to the right of the Cube drop-down list to filter the list of dimensions by cube, and to see which cubes are marked as Dense.
3. Select a sparse dimension for the attribute.

   Note:

   Only sparse dimensions can contain attributes.

   See About Sparse and Dense Dimensions.

4. Select the Edit Dimension Properties tab .
5. Select options:

   • To create an attribute, next to Custom Attributes, click Create. Type an attribute name, and select a data type: Text, Date, Boolean, or Numeric.

     Note:

     • You can't modify the type after the attribute is created.

     • Before working with date attributes, you must select the Attribute Dimension Date Format in Application Settings and save it.

     See Understanding Attribute Data Types.

   • To modify attributes, under Custom Attributes, click (Actions) next to the attribute you want to modify, select Edit, and then update the attribute name.

6. Click Close.

   When you click Close, the hierarchy is validated and an error displays if issues are detected. For example, date attribute values must be entered in the correct format, and numeric and date attribute dimensions must have at least one attribute value defined.

7. Update and validate business rules and reports.

Tip:

Instead of explicitly filtering by an attribute (like Red, for instance), you can create a user variable for the attribute dimension, and then use the user variable as the filter. Then you can enable the user variable as a dynamic user variable which would allow users to change the value of the filter at runtime. This is a useful technique that allows for dynamic filtering.

See Managing User Variables.