Based on the previous data discovery stages (Data Profiling, relationship detection, datasource logical data model) we have now to rework the data by managing at least the reject based on the non-matches joins (see the logical Data Model). We’ll have to decide with the business what to do with this dataset, for example:
- Removing them by filtering the initial datasource
- Altering the data by creating for example a surrogate data (a new Identifier in which the non match will join)