You might perform exception management towards the end of a data quality project. For example, you may run a deduplication and consolidation operation on a data set to identify and resolve duplicate records. However, you may be interested in the status of the records whose duplicate scores are on the border of the deduplicate threshold that the operation defined. You can configure a rule specification to identify the records with borderline scores and to add exception indicators to the records.

Or, you may complete a pattern-based parsing operation and find that one or more key data columns still contain null data. You can configure a rule specification to flag the records that contain the null values and to add exception indicators to the records.

In each case, the rule specification adds data columns to the records to store the exception indicators. You can use a profile to review the records.

In the exception management process, you create an exception task from the profiling task that applies the rule specification to the data. At run time, the exception task selects the records that contain the exception indicators and writes the records to a target that you can download for further processing.

You may decide to process the data further in Informatica Intelligent Cloud Services. For example, you might run additional deduplication and consolidation operations on the data set and apply a lower deduplication threshold. Or, you may decide to manually review and update the records in another application.