Start with the practice CSV, retain identifiers as text, preview two formatting rules, and export a checked result. Product identity and conflicting prices remain review decisions.
In this guide
Prepare your computer and a practice file
Choose the official download for your operating system and follow its setup instructions, including Java requirements for that package. Launch the application and keep its process running while using its browser interface.
Complete your first import and cleanup
- Launch OpenRefine and use Create Project to select the practice CSV.
- Inspect the preview: seven columns, correct quoting, and identifiers preserved as text. Disable automatic number parsing for this exercise.
- Create the project and check 12 rows. On product_name use Edit cells → Transform to trim boundary whitespace; do the category rule separately.
- Follow the detailed workflow, inspect exceptions, and clear filters before a full CSV export.
Check the export and save the history
If leading zeros have disappeared, reimport the untouched CSV with the correct types. If the export has too few rows, inspect active filters. If a rule affects unexpected values, use history to return to the earlier state before applying another rule.
- The output retains all 12 rows, SKUs, prices, and quantities.
- The missing category and conflicting pair still need a decision.
- Use Undo/Redo → Extract to save the reviewed formatting operations for a matching later file.
Where it runs and what AI adds
Basic cleanup runs in the local application. These two formatting rules need no AI account. External services or extensions are separate choices; inspect what you enable before adding them.
References
Use it in a task
Product Spreadsheet Cleanup for Store Owners: Clean, Check, Repeat
Continue the task
Before reusing a cleanup routine, check the source values and the rules that determine record identity.