I have 3 big fact tables in the source which i load incrementally. Primary keys are CompanyID and EntryNo (applicable for each table separately, incrementing), the records in these fact tables are never deleted.
I need to merge these tables to a single table in DSA layer also incrementally. I think of following setup:
- Make the DSA table also incremental, check “Don’t handle deletes” (as the records are never deleted in source)
- Include both CompanyID and EntryNo as primary keys in the table;
- Incremental Selection Rule → EntryNo > (Last max value);
- Map all tables from source in this single DSA table

The result which i expect is every time DSA table is executed - it incrementally loads data from 3 related fact tables - based on EntryNo for each separate CompanyID (one CompanyID per one source table). This way i have one huge table containing all the records for all 3 “Companies”. Am i correct to assume this is the setup i need? Are there any risks with possibly missing or overwriting the data?
Thank you