Because it compresses the data

Because it records the clean-up as an ordered recipe that replays on new data

Because it edits the source file so the fix is permanent

Because it removes duplicates automatically