True, there is about 1,000+ records with Parcel IDs in GIS_DATA (some of these are duplicate IDs, as mentioned). Field_Data gets populated with Parcel IDs once users add field data to the record. Currently, there are 10 records in this table. I use an identity column as primary keys in both tables.

Is there a way to structure the tables/relationships so I don't have clean up the data?

I could make the ParcelID in the FieldData table a primary key?

Any other advice on my archectecture?

Thank you,

Paul