auto-fix by padding assumes that the only way to get an ID that's too short is by excel removing the leading zeros - presumably it could also be human error whe

auto-fix by padding assumes that the only way to get an ID that's too short is by excel removing the leading zeros - presumably it could also be human error when entering etc. (in theory I suppose that would be validated in whatever form the number is entered into in the app that generates the CSV, but hard to say for sure). Or some other issue in data processing pipeline erroneously truncating the ID number some how. Validating the data would at least prevent getting invalid data into database (and presumably this is happening already), but it doesn't actually "fix" the problem, you still then need the original provider of the data to fix what's missing.