Thanks for the prompt response Guust. The data volumes are going to be in the tens of thousands so I need to make it as automated as possible, and avoid an export if I can. However, your comments around “post ID” suggest that it’s the primary key for the listing and is used by the csv import system for record matching on the import, where as I would need to define the key myself to follow the logic I outlined in the original question.
An alternative approach would to be create a “feed ID” and then:
Start
Delete all records where feed ID = X and region = Y
import feed data as new records for region = Y, and feed ID X
End.
That’s less efficient,but would work. Is that possible?