Yeah because SCA doesn’t have easy access to subrecords, I’m flattening them into a text format in a user event script, which works fine. But when a csv import is run, the script is triggered, sees no subrecords, and blanks out the field.