Another type of typical data quality error involves columns that should only contain data if certain data values (or null) are present elsewhere. For example, in the customer table where some customers are organizations and others are individuals, only organizations will have SIC code, organization size, etc. (unless the presence of that data actually represents data for the individual's organization).
Typically the handling of these potential violations involves nothing more than a post-ETL report on just-loaded data that creates visibility into the situation. These reports are not useful unless they are reviewed from the appropriate systems and business personnel.
Read part 1.
For more information, check out this data quality guide.
