OpenRefine: clean a catalogue without hiding the changes

OpenRefine is an open-source tool for working with messy data, including cleaning, transformation and enrichment through external services. It is useful for a catalogue assembled from several exports, where inconsistent spellings, empty fields or repeated values make search and comparison unreliable. Keep the original dataset unchanged and work on a copy. Inspect a few columns, group suspicious values and decide which differences are errors and which preserve meaningful distinctions. Make small transformations, review their effects and keep a record of the decisions. Test the cleaned export in the system that will consume it, especially for dates, identifiers and non-English names. Similar text does not always describe the same entity. Automatic grouping can merge different people or collapse historical terminology that matters to the research. External reconciliation services introduce their own coverage and matching assumptions. Review proposed matches before accepting them, preserve stable original identifiers, and document the rules so another editor can reproduce or challenge the cleanup.

Editorial review: 2026-09-23. New evidence may change this account.

Report a problem with this record