Integrations
OpenAlex data flows into a lot of other tools. This page covers exporting from the website and the most common third-party integrations. For programmatic access, see Get the data.
How do I export results from the OpenAlex website?
Above any works result set there’s an export button; exporting needs a free OpenAlex account. Pick a format:
- CSV (Excel) and CSV (.csv) — for Excel and friends; flattens the works data into the columns you pick. Both cut any cell longer than 30,000 characters, under Excel’s limit of 32,767, so a very long list such as a work’s references loses its tail. CSV (Excel) also puts each work on one line (a line break inside a value, as in some raw affiliation strings, becomes a space) and starts with a byte-order mark so Excel shows accented names intact. Plain CSV keeps those line breaks inside quoted cells, which pandas and R read as one row per work.
- EndNote (RIS) (.ris) — for reference managers like EndNote, Zotero or Mendeley.
- BibTeX (.bib) — for LaTeX and BibTeX-based reference managers.
- Text (.txt) — the Web of Science plain-text format (tags such as
AU,TI,TC; each record ends withER), for bibliometric tools that read Web of Science files. Each record’sUTline holds its OpenAlex ID.
The export draws on your account’s daily API budget, and the dialog shows what it will cost before you start. The system prepares the file and gives you a download link — large result sets can take several minutes.
How do I download an OQL table?
An OQL table of groups, such as get the count and percent open access of [Kenya](KE)'s works by year since 2015, downloads from the same button above the results. It exports every row, however many there are: one CSV with no split, or a zip with two splits (groups.csv, one row per group, plus a summary of the whole set). The Export dialog shows the price before you start: the query’s price for every 100 rows. On the API, add &format=csv to the query for the groups, and &format=csv&table=summary for the summary. Details: Downloading results.
How do I import OpenAlex data into VOSviewer?
VOSviewer builds and visualizes bibliometric networks, and it talks to OpenAlex directly — no file downloads needed:
- At openalex.org, build the query you want to analyze. Above the results, click the share icon and choose Copy query as… → API URL.
- In VOSviewer, click “Create” → “Create a map based on bibliographic data” → “Download through the API” (OpenAlex is the default).
- Paste your URL into the “Request URL” field and hit next to specify your network.
VOSviewer reads only the search and filter parts of the URL. A query with nested AND/OR groups has no filter= form, so its API URL comes out as an ?oql= link that VOSviewer can’t use; keep a VOSviewer query to plain filters and a single search.
Use VOSviewer 1.6.21 or later and give it your free OpenAlex API key (get one here): downloading past the first page of results needs a key, and 1.6.21 is the first version that can send one.
Note that VOSviewer supports up to 50,000 records per analysis, so you may need to narrow your query with extra filters.
Link-resolver integrations
Several link resolvers — SFX, EBSCO Discovery Service, 360 Link — use Unpaywall (OpenAlex’s legacy-format API surface, same underlying data) to route users to open-access copies: OA copies → Unpaywall data → link resolver → your users. They use one of two free services: a redirect service (unpaywall.org/<DOI> sends you to the best OA location, or the publisher page if there isn’t one) and the Unpaywall REST API, which returns the best OA location as metadata instead of redirecting.
When something breaks, the split matters: we can help when the problem is the link between a DOI and an OA location in our data; your link resolver vendor is the right contact for everything else (installation, formatting, display). To triage:
- The link looks like
unpaywall.org/<DOI>(the redirect service): if it lands on a toll-access publisher page, that’s working as designed — we just don’t have an OA copy. If it lands on the wrong article or an incomplete copy, that’s ours to fix — report it with the DOI. - It’s a direct link to a copy (the API): these should only appear for OA articles, so anything other than a full, free copy of the right article is likely our bug — let us know.
Why does the Unpaywall extension say a paywalled article is open access?
Besides using the API, the browser extension looks for a PDF link directly in the page you’re viewing and, if it finds one, shows a Bronze OA icon. If you’re on a university network or logged into a publisher site that gives you access, the extension will happily point at that PDF — even though the article isn’t actually open access. That’s deliberate: the extension is built to help you obtain fulltext, not to educate about access status. If the article’s true OA status matters to you, check the API directly. And if the link it gives you doesn’t work, report it.
Last updated
View as Markdown