About us
Turning data into impact narratives.
About openscience.works
openscience.works builds open, reusable services that help researchers, libraries, and publishers understand how open scholarly works are used, shared, and discussed across the research ecosystem. Drawing on open data and transparent methods, the platform turns diverse signals into clear, contextual impact narratives rather than opaque scores.
Instead of reducing research to single metrics, openscience.works focuses on helping organisations tell richer, evidence-based stories about the reach and value of open scholarship. The services are designed to support responsible, discipline-sensitive interpretation and to make every data source transparent and inspectable.
Methodology & Data Sources
How do we calculate impact?
We believe that citations alone do not tell the full story of a piece of research. This engine looks at a variety of signals to determine the impact of a work, resulting in what we call Inferred Roles.
Inferred Roles & Classification
Our pipeline goes beyond simple citation counts. By analyzing the composition of a work's attention, usage, and citing landscape, we dynamically assign "Inferred Roles" across four major impact domains:
1. Academic & Scientific Impact
Scholarly Uptake
High citation density within core academic journals and monographs. The work serves as a foundational building block for further research.
Rapid Uptake
The work has accumulated a high volume of citations or downloads in an unusually short period since publication, indicating immediate relevance and urgency within the field.
Reference Point for Synthesis
Frequently cited in literature reviews, meta-analyses, or encyclopedic entries, serving as a "shorthand" for a specific finding or historical overview.
Methodological Anchor
The work serves as a standard protocol, framework, or software tool utilized by other researchers (triggered by specific citation contexts or high usage in datasets and software).
Evidence-Bearing Reference
The work is explicitly applied as hard data or methodology (triggered by high "supporting" citation tallies via scite or links to registered clinical trials).
2. Readership & Educational Impact
Usage-Driven Uptake
The work demonstrates immense value through direct consumption rather than formal citation (triggered by high download counts on platforms like OAPEN/OPERAS, massive HTML views, or heavy library holdings).
Pedagogical Anchor
The work has been widely adopted for teaching and training (triggered by inclusion in Open Educational Resources (OER) syllabi, reading lists, or textbook citations).
3. Public Engagement & Media
Active Public Discourse
High, fast-moving engagement in community spaces; the work has sparked conversation (triggered by heavy activity on X/Twitter, Reddit, Facebook, or community blogs).
Public Visibility & Knowledge Base
The work serves as a trusted reference for the general public, establishing a lasting public record (triggered by citations in Wikipedia, Stack Exchange, or non-academic wikis).
High-Visibility Uptake (Media)
The work has "broken out" into mainstream awareness (triggered by mentions in prestigious global news outlets, broadsheet newspapers, or high-impact professional magazines).
4. Practical & Real-World Application
Sustainability & Policy Relevance
Direct influence on governance and societal frameworks (triggered by citations in government white papers, WHO reports, NGO briefs, or UN policy documents).
Commercial Linkage
The work crosses over into the retail or commercial sphere, indicating consumer interest or industry application (triggered by strong presence on Amazon, Goodreads, or patents).
Intelligence Markers & Badges
Our reports use dynamic visual chips and status badges in the header to provide an immediate census of a work's output type, integrity, and real-world application.
Work Type & Access
Defines the specific form and availability of the scholarly contribution:
Status & Integrity Badges
Dynamic tags that signal publication maturity and research integrity:
- Peer-Reviewed & Published ↗ Identifies a preprint that has successfully passed peer review and been formally published.
- Early Sharing / Preprint ↗ Links a published journal article back to its original preprint version.
- Published Correction Highlights official publisher updates, errata, or addenda tracked natively via Crossmark.
- Retracted Signals severe issues detected via the Retraction Watch database.
Impact & Context Badges
Badges that track how research data is shared, applied, and discussed globally:
- Methodologically Supported Powered by scite.ai, indicates the work is frequently cited for its methods or foundational frameworks.
- Linked to Data Highlights linked datasets and software repositories (Zenodo, Figshare, Dryad).
- Proven Reuse Tracks downstream reuse of datasets and software by other publications.
- Clinical Trial Links the publication to a registered clinical trial found via Europe PMC.
- Community Commentary Highlights active post-publication peer review and methodological discussions via PubPeer.
Data Sources & Integrations
Traditional academic metrics only tell a fraction of the story. To build a comprehensive picture of a work's true impact, OpenScience.works aggregates real-time data from a diverse ecosystem of global APIs, repositories, and custom heuristic engines:
- Academic Impact & Core Metadata: Powered natively by OpenAlex, Crossref, and DataCite to capture global citation networks, geographic distribution, institutional affiliations, and UN Sustainable Development Goal alignment.
- Educational & Teaching Adoption: We query Open Library and WorldCat for community reading lists, scan OpenCourseWare and the Open Textbook Library for formal syllabi, and actively mine YouTube transcripts to verify when a work is assigned in recorded university lectures and symposiums.
- Research Integrity & Scrutiny: We surface severe breaches via the Retraction Watch Database, track active post-publication peer review through PubPeer, and utilize Crossref's Crossmark data to highlight routine scientific self-correction (official errata, corrigenda, and addenda).
- Stewardship & Reproducibility: Through Europe PMC and DataCite, we deep-mine full-text literature to find linked data repositories (Zenodo, Figshare, Dryad), funding grants, Research Resource Identifiers (RRIDs), and registered Clinical Trials. We also surface community peer reviews for preprints via Sciety.
- Open Access & Usage: We cross-reference platforms like Thoth Open Metadata and DOAB for OA verification. Download metrics are captured via the OPERAS Metrics API and OAPEN, while commercial readership is assessed through Google Books, Goodreads, and Amazon.
- Public Discourse & Altmetrics: To measure how research permeates society, we track organic mentions across the open web. We are actively migrating to direct API fetchers to capture discussions on Wikipedia, mainstream news outlets, scientific blogs, Reddit, Bluesky, StackExchange, and open web annotations via Hypothesis.
- Citation Context: We integrate with scite.ai to look beyond simple citation counts, analyzing the full text of citing papers to determine whether a work is being "mentioned," "supported," or "contrasted."
How We Calculate Confidence
Because some of our metrics rely on text-mining rather than strict DOI matching (such as our YouTube lecture tracker or OpenCourseWare scanner), we employ an AI Academic Confidence Score. This score evaluates the context of a mention—for example, verifying if a YouTube channel belongs to an academic institution and checking the transcript for specific educational keywords—ensuring we filter out casual mentions and noise.
About Martijn Roelandse
Martijn has more than 20 years of experience in scholarly publishing and biomedical research. After completing his PhD in Basel, he built a career at the intersection of publishing, technology, and open science, with a particular focus on books and journals.
In 2014 he was co-founder of Bookmetrix, a Springer–Altmetric initiative that pioneered book- and chapter-level metrics by aggregating multiple signals of reach, attention, and use. Through this work, Martijn gained first-hand experience in the opportunities and limitations of impact reporting, including the importance of transparent data provenance and discipline-sensitive interpretation, especially in the humanities and social sciences.
A seasoned editor and publishing professional, Martijn brings deep knowledge of the scholarly communication ecosystem and specialises in building bridges between startups, publishers, libraries, and research organisations. At openscience.works, he focuses on partnerships, product direction, and ensuring that tools are aligned with real-world publishing and library workflows.