A financial services company needs to implement a data pipeline for transactional data, which must be cleansed and validated for compliance purposes before loading into BigQuery. They also plan to conduct further analytics on the data after loading. Which data manipulation methodology is the most appropriate for these requirements?