For researchers who cite data
Paste a DOI or a bad export. See exactly what's wrong, get a correct citation in seven styles, and export clean BibTeX, RIS or CSL JSON.
Paste a DOI or repository link
Try one
10.5281/zenodo.3678171
https://doi.org/10.5061/dryad.j6q573nb3
How it works
Drop in a DOI or repository link, a broken BibTeX / RIS / CSL export from your reference manager, or build a record manually for data shared without a DOI.
Citable checks the record against the DataCite or Crossref registry, fixes the resource type, version, authors and year, and shows you exactly which fields changed — with a citation health score.
Copy a journal-ready citation in seven styles, download clean BibLaTeX, RIS or CSL JSON, and generate a Data Availability Statement or Zotero Extra snippet.
FAQ
Most reference managers were built for journal articles and books. When you import a dataset, they often record it as a journal article, drop the version, or lose the publisher (the repository), which produces citations that don't follow DataCite or FORCE11 data citation principles. Citable detects those errors and repairs them against the official registry record.
Seven styles: APA 7, Chicago, MLA 9, Harvard, IEEE, Vancouver and Turabian. You can switch between them instantly and copy or download the result.
Use the “No DOI” tab. You can build a citation manually with authors, title, year, version, publisher and any identifier you have (URL, Handle, ARK, accession number), including an accessed date for informally shared data.
Clean BibLaTeX, RIS and CSL JSON, plus a Zotero “Extra” field snippet that forces the dataset type and version even in stubborn libraries. Batch mode exports a whole cleaned reference list at once.
No. Everything runs in your browser. Metadata is fetched from the public DataCite and Crossref APIs, and nothing you paste is ever uploaded or saved.
Yes. Citable is a free microtool for academic researchers — no account, no sign-in, no limits.
Suggest a new useful feature
Citable is a microtool on purpose — it does one job well. If there's a dataset citation problem it doesn't solve yet, tell us and we'll consider adding it.