Data Correction with Data Quality Services - SQL Server Integration Services Design Patterns - page 163

5. Repeat steps 3-4 for each field you will be cleansing. The Lookup
transforms should be connected using the Lookup Match outputs.
6. Add a Conditional Split transform with a single expression that
checks each of the flag fields. If any of the flag fields are NULL , the
row should be sent to DQS for proper cleansing. For example, the ex-
pression to check the RegionCodeFlag for a NULL value would
be ISNULL([RegionCodeFlag]) .
7. Connect the Conditional Split output you created to the DQS Cleans-
ing transform. Rows going to the Conditional Split's default output
can be ignored (since their values were successfully validated using
the Lookup transforms).
8. Complete the rest of the data flow based on the appropriate logic for
handling the output of the DQS Cleansing transform.
Figure 5-14 shows a screenshot of a data flow to cleanse a single field using the
preceding logic.