Tenzing is a free, open-source tool that helps research teams turn a simple contributor spreadsheet into a standardized CRediT (Contributor Roles Taxonomy) statement, ready to drop into a manuscript submission or hand off to a publisher’s production system. It exists specifically to solve the mechanical problem CRediT creates once a team adopts it: someone still has to collect who-did-what across every co-author, map it consistently onto CRediT’s fixed set of roles, and format the result the way a given journal wants it. Tenzing automates that step rather than replacing the taxonomy itself.
What Tenzing is, and who built it
Tenzing is a web application and companion R package, described in a peer-reviewed methods paper, “Documenting contributions to scholarly articles using CRediT and tenzing” (Holcombe, Kovacs, Aust & Aczel, PLOS ONE, December 2020). The authors are affiliated with the University of Sydney, ELTE Eötvös Loránd University (Budapest), the University of Cologne, and the University of Amsterdam, and the project is maintained under the “Contributorship Collaboration” banner, whose stated goal is making it easier and more transparent to curate and report scholarly contributions.
The name is a deliberate reference to Tenzing Norgay, the Sherpa mountaineer who reached the summit of Everest alongside Edmund Hillary in 1953 but has historically received less individual credit for it than his climbing partner — a fitting namesake for a tool built to make sure every contributor’s role is recorded explicitly rather than left to informal convention.
How Tenzing works
Tenzing does not ask contributors to fill out a form one at a time. Instead, it works from a single spreadsheet that lists every team member as a row, with a column for each of CRediT’s 14 contributor roles (Conceptualization, Data Curation, Formal Analysis, Funding Acquisition, Investigation, Methodology, Project Administration, Resources, Software, Supervision, Validation, Visualization, Writing – Original Draft, and Writing – Review & Editing):
- Start from the template. The project provides a ready-made Google Sheet / Excel template with a row per author and a checkbox-style column per CRediT role, plus an optional “degree of contribution” indicator (lead, equal, or supporting) for each role an author held.
- Fill it in as a team. Because it’s one shared file rather than 14 separate emails, a corresponding author or a delegated coordinator can populate it collaboratively, then confirm it with each co-author before submission.
- Upload it to the app. The completed sheet is exported as CSV, TSV, or XLSX and uploaded to the Tenzing web app at tenzing.club (or processed locally via the R package).
- Generate the outputs. Tenzing parses the file and produces several ready-to-use outputs in one pass rather than requiring the role mapping to be redone by hand for each format a journal or publisher wants.
What Tenzing generates
From the same input spreadsheet, Tenzing can produce:
- A plain-text CRediT contribution statement formatted for a manuscript’s author-contributions or acknowledgments section.
- A formatted author/affiliation block for a manuscript’s first page.
- JATS-XML markup of the contributorship data, suitable for a publisher to ingest directly into article metadata.
- A YAML block compatible with the
papajaR Markdown package, for teams writing APA-formatted manuscripts in R. - Supporting sections such as a funding-acknowledgment statement and conflict-of-interest text, drawn from the same underlying spreadsheet.
The point of generating several formats from one source file is consistency: the CRediT statement in the manuscript body, the structured metadata a publisher stores, and any supplementary contributorship record all trace back to the same set of checkbox entries, rather than being retyped separately and risking drift between them.
Web app vs. R package
Tenzing is distributed two ways, and which one a team uses is mostly a matter of workflow preference rather than a difference in output:
- Web app (tenzing.club, hosted as a Shiny app): no installation required — upload the spreadsheet in a browser and download the generated files. This is the more accessible route for a corresponding author who just needs the statement once, at submission time.
- R package (source on GitHub at
marton-balazs-kovacs/tenzing): installable from GitHub and run locally withtenzing::run_app(), or used non-interactively as part of a scripted or reproducible-manuscript pipeline — for example, regenerating the CRediT block automatically each time an R Markdown or Quarto manuscript is rendered, so the statement can’t silently fall out of sync with the underlying spreadsheet.
Tenzing’s source code is released under an MIT license (with an additional license file in the repository covering specific components), and a snapshot of the codebase is archived on Zenodo, giving it a citable, versioned DOI independent of the live GitHub repository — useful for a methods section that wants to cite the exact tool version used.
Where Tenzing fits relative to CRediT itself
Tenzing is a documentation and formatting aid, not a governance body or a replacement for editorial judgment. It does not decide who qualifies for a given CRediT role — that determination is still made by the contributors and the corresponding author, guided by the taxonomy’s own role definitions (see CASRAI’s CRediT & authorship attribution overview) and, where relevant, a journal’s authorship policy. It also does not adjudicate authorship eligibility in the ICMJE sense — a CRediT role and ICMJE authorship criteria answer related but distinct questions, covered in more depth in CASRAI’s comparison of CRediT vs. ICMJE authorship.
Teams weighing whether an author can hold several CRediT roles, or whether a role can be split between two people, should see How many CRediT roles can one author hold, and can a role be shared? — Tenzing’s spreadsheet format directly supports both cases through its per-role, per-author checkbox structure. For teams that need to disclose generative-AI assistance alongside human contributor roles, see how to disclose AI assistance in a CRediT statement, role by role; Tenzing’s output is plain text and JATS-XML, so an AI-disclosure line can be appended to the generated statement using the same conventions. Teams wanting worked examples of finished statements in different formats before building their own spreadsheet can also see CASRAI’s CRediT author statement samples.
Limitations to know before adopting it
Tenzing is a formatting and consolidation tool, not a validation service — it will faithfully generate a statement from whatever roles are checked in the spreadsheet, so the accuracy of the final statement still depends entirely on the team completing the input file honestly and completely. It also does not integrate directly with a journal’s submission system; the outputs are designed to be copy-pasted or attached, not auto-submitted. As with any actively maintained open-source project, teams should check the current GitHub repository for the latest supported input formats and any changes to the CRediT role list before relying on a specific version for a submission.
Frequently asked questions
Is Tenzing free to use?
Yes. Tenzing is open-source software (MIT-licensed) and the hosted web app at tenzing.club is free to use; no account or payment is required to generate a contribution statement.
Does a journal need to specifically support Tenzing to accept its output?
No. Tenzing’s plain-text and formatted outputs are designed to be pasted into a standard author-contributions section, so any journal that accepts a CRediT statement in its normal submission workflow can use Tenzing’s output regardless of whether the journal has any specific integration with the tool.
Can Tenzing assign or validate CRediT roles automatically?
No. Tenzing consolidates and formats roles that the contributors themselves select in the input spreadsheet; it does not determine eligibility for authorship or for any individual CRediT role.
Does Tenzing require every co-author to use the tool themselves?
No. Typically one coordinator (often the corresponding author) populates the shared spreadsheet and confirms the entries with each co-author, then runs the file through the web app or R package once to generate the outputs.
