When importing customer data via CSV, duplicate records are created if the file contains repeated entries. The import script lacks idempotency checks against existing records. Implementing a deduplication step based on unique identifiers will prevent redundant data.