In Phase 1 of PDH, numerous redundant attributes and attribute groups were created, leading to millions of unnecessary data entries that required cleanup. The key challenges included the inability to delete attributes with existing data using standard PDH functionality, the need to transfer data between character and number groups and the inefficiency of the existing process, which took weeks to prepare a new instance due to the large data volume.