Development and implementation of an automated process designed to identify suspect data and prevent the corruption of a business critical database.
Poor quality data can lead to poor quality service and poor quality decisions. In some systems it is critical that erroneous data is identified and prevented from entering the system. One way of achieving this is to set up automated procedures that will check the quality of an input source and look for patterns that may indicate a set up issue of underlying data quality issue for example:-
- First name and surname are presented the wrong way round
- The supposed Date of Birth field is actually an account open date
- The initials field is repeating the first character of the first name
- There are too few / too many house names present
There are a great many common data errors that can be identified provided you know what “normal” looks like.
- Data quality can be a very grey area
- People often think their data is fine and may not take kindly to being shown that it isn’t.