A

Recommendations for Revising the Open Access Classification System in OpenAlex

Zenodo (CERN European Organization for Nuclear Research)

Abstract

The open access (OA) color classification system used by Unpaywall (i.e., Gold, Green, Hybrid, Bronze, and closed) was introduced in 2018 and has since become widely used across the ecosystem, including by Scopus, Web of Science, Dimensions, and numerous funders and institutions worldwide. However, the scholarly communication landscape has changed significantly since these categories were defined and the limits of this system are proving problematic. In particular, the current system struggles with research outputs that are not journal publications, relies too heavily on journal-level policies rather than work-level attributes, conflates business models with access status, and is difficult to accurately operationalize in an environment where journals regularly change the OA status of publications en masse. The OpenAlex Colors Working Group was convened by the OpenAlex Community Advisory Board in January 2026 to evaluate these limitations and recommend a revised framework. After consultation and analysis, the Working Group recommends replacing the color-based system with a three-dimensional classification framework based on: Access — whether a work is freely accessible to read or is closed Location — whether a work is available at the publisher or on another platform (repository, preprint server, etc.) License — whether a work carries an open license, a closed/restrictive license, or has no clearly identifiable license This framework enables descriptive, work-level labels (such as Publisher Open, Platform Public, or Closed) that can accommodate articles, preprints, datasets, dissertations, books, and other scholarly output types, while avoiding the confusion that comes with the previous classification. Under the proposed system, venue-level classifications are left to the user, but can be derived from observed aggregate behavior, or left up to a separate classification system (see recommendations). Importantly, this proposal focuses on access, location, and license but does not address the business models of publishing venues (e.g., those funded with Article Processing Charges, Hybrid, diamond, subscribe-to-open, transformative agreements). One of the key complications of the existing color system is that it blends access status (a property of an article) with business models (a property of a venue)—distinct dimensions that cannot be easily reconciled. The Working Group recognizes that business model classification is an important and distinct need, and recommends that a separate initiative be established to develop a complementary venue business model taxonomy. This report is intended to capture the discussion of the Working Group. In recognition that any eventual implementation in OpenAlex will impact the broader community, the Working Group invites feedback from funders, institutions, publishers, librarians, database providers, and researchers ahead of making their final recommendations for consideration for implementation by OpenAlex.

Authors 10

  1. Simon Fraser University · Public Knowledge Project

    Affiliation as printed

    Simon Fraser University

    Public Knowledge Project

    Scholarly Communications Lab

  2. Affiliation as printed

    Sesame Open Science

  3. Gates Foundation

    Affiliation as printed

    Bill & Melinda Gates Foundation

  4. Sorbonne Université

    Affiliation as printed

    Sorbonne Université

  5. Leiden University

    Affiliation as printed

    Leiden University

  6. Directory of Open Access Journals

    Affiliation as printed

    Directory of Open Access Journals

  7. Scientific Electronic Library Online

    Affiliation as printed

    Scientific Electronic Library Online

  8. European Bioinformatics Institute

    Affiliation as printed

    European Bioinformatics Institute

  9. University of Cambridge

    Affiliation as printed

    University of Cambridge Trinity College

Cited by 0 stored of 0

No patents citing this paper on Lens.org (checked 2026-10-11).

References 0