You can include more than one dimension in a data classification, so that Data Archive uses all the dimensions to create the segments. When you apply multidimensional data classifications, the ILM Engine creates segments for each combination of values.

For example, you want to create segments for a segmentation group by both a date range and sales office region. The application has three years of data that you want to create segments for. You choose the time dimension to create segments by year. You also create a custom dimension called region. You configure the region dimension to create segments based on whether each sales office is in the Eastern or Western sales territory.

When you run the segmentation policy, the ILM Engine creates seven segments, one for each year of data from the Western region, one for each year of data from the Eastern region, and a default segment. Each non-default segment contains the combination of one year of data from either the East or West sales offices. All remaining data is placed in the default segment. The remaining data includes data that does not meet the requirements for a segment, such as transactions that are still active, plus all new transactions.