Skip to main content

OpenRefine

Editor’s pick

Cleaning tabular data

Free, open-source data cleaning tool for tidying messy tables and standardising text in bulk.

Main category
Data cleaning
Cost
Free
Licence type
Open source
Platforms
Windows / macOS / Linux
Visit OpenRefine Quick start Code repository

Version-watched: add it to favourites and confirmed new releases will show on your favourites page (stored in this browser only).

Editor’s pick for System & files · Cleaning tabular data

Good for
Anyone who needs to unify inconsistently spelled categories in a table or tidy text in bulk.
Why it is recommended
Runs as a small web server on your own computer and needs no internet for basic functions; facets, clustering and bulk transformations find and merge differently written versions of the same category.
Limits and when to choose something else
Needs Java (a Windows package with Java is available); default memory is 1 GB and must be raised for large tables; it is not a statistics package; reconciliation sends data to the external service you use.

Sources checked: 2026-11-05 · Compiled from official sources; not run by ZIRULAB

What it is for and who it suits

Anyone with hand-entered or merged tables who needs to unify inconsistently spelled categories, trim spaces, or split and join columns.

How it is used · Where data is processed

How it is used
Graphical interface
Where data is processed
On your computer

Runs as a small web server on your computer and is used in the browser; basic functions need no internet. Reconciliation and similar features send data to the external service you choose.

Getting started: Unify inconsistently written categories

Steps compiled from the official documentation and interface; not run by ZIRULAB. Practise with demo data.

  1. Prepare demo data: a column of species or place names with deliberate case, spacing and spelling differences (e.g. “Pinus densiflora”, “pinus densiflora”).
  2. Create a project in OpenRefine and import the table; on that column choose “Facet → Text facet” to see each variant and its count.
  3. Click “Cluster” in the facet, review each suggested group and merge only those you are sure about.
  4. Check the facet again, export the cleaned table, and save or export the operation history so the cleaning can be repeated.

Frequently asked questions

Do I need Java?
Yes, Java is required; the Mac version and one Windows package include it. See the official installation page for versions.
Is my data uploaded?
OpenRefine runs on your computer and basic functions need no internet; it goes online only to import from the web, reconcile with web services or export to the web.
What if a large table will not open?
The default memory is 1 GB; raise it as described officially (generally up to half of the free memory).

Details

OpenRefine is a powerful free, open-source data cleaning tool for handling messy tables, duplicate names, inconsistently formatted fields, bulk text transformations and data completion. It is better suited than Excel to building reproducible data cleaning workflows, and can connect to external data services to enrich data.

Use cases: table cleaning, species name standardisation, place name tidying, literature data organisation, bulk text processing

Choosing between similar tools

  • LibreOffice: For simple sorting, filtering and formulas, a spreadsheet such as LibreOffice Calc is enough.
  • R / RStudio: Use R when cleaning should live in the same reproducible script as the analysis.

Used in these research workflows

Sources and scope of the check

Checked 2026-11-05 · Compiled from official sources; not run by ZIRULAB