Glossary

Data cleaning

Turning raw records into a dataset you can trust, by applying stated rules and keeping evidence of every change.

Data cleaning applies explicit rules to raw data: trimming, type conversion, validation, deduplication, joins and reshaping. A good cleaning process never edits the raw input, records why each row was changed or rejected, and produces the same result when run again on the same data.

Practise: Clean an order-line export