Skip to content

Data Licences Explained

For people

CC0, CC BY, CC BY-SA, ODbL, non-commercial terms and government licences: what common data licences allow.

Editorial team 2 min read

Share X LinkedIn Reddit Bluesky Email

Before using a dataset, check its licence. Being able to download data doesn't mean you can use it any way you like.

Common Licences

Licence In practice
Public domain / CC0 No conditions; use freely
CC BY 4.0 Any use, including commercial, with attribution
CC BY-SA 4.0 As CC BY, and adaptations you share must use the same licence
ODbL Open Database Licence: attribution and share-alike for databases
CC BY-NC Non-commercial use only
Custom terms Read carefully; may restrict use or redistribution

Many government datasets use their own open licences, often similar to CC BY.

Attribution

Licences requiring attribution usually expect the creator's name, the dataset title, a link to the source and the licence. Dataset cards often give a preferred citation.

Share-Alike

Share-alike licences apply to adaptations you distribute — combined datasets, for example. Think about this before mixing share-alike data into a product.

Non-Commercial Restrictions

"Non-commercial" can be ambiguous. If there's any doubt about whether your use counts, ask the publisher.

No Licence Means No Permission

Absence of a licence doesn't make data free to use. Contact the owner.

Personal Data

A licence covers copyright and database rights; it doesn't override privacy law. Data about people may carry obligations regardless of its licence.

Keep Records

Log the source, version, licence and attribution for every dataset in your projects. This is general information, not legal advice.

More in Data for AI

All Data for AI guides →
Data for AI Guide · 1 min

How to Read a Dataset Card

The questions a dataset card should answer — what one row is, where the data came from, its licence and its quirks — before you use it.

Data for AI 1 min read 9 Apr 2026 For people

Data for AI Guide · 2 min

Data Quality Dimensions

Accuracy, completeness, consistency, timeliness, validity and uniqueness: a framework for checking whether data is fit for purpose.

Data for AI 2 min read 8 Apr 2026 For people

Data for AI Guide · 2 min

Handling Missing Data

Why data goes missing, how to find out, and the options — dropping, imputing, flagging — with their trade-offs.

Data for AI 2 min read 7 Apr 2026 For people

Data for AI Guide · 2 min

Detecting and Handling Outliers

How to spot unusual values, decide whether they are errors or genuine extremes, and treat them appropriately.

Data for AI 2 min read 6 Apr 2026 For people

Discussion 0

Use @handle to bring somebody in, and #tag to file it.

Nothing here yet. Ask a question, or say how you used it.

Sign in to join the discussion. Sign in