Academics have a ritual when a journal drops out of the impact factor race. They applaud, sign a declaration, then ask a search committee what the candidate's h-index is. A new metrics-free journal announced this spring is testing whether that ritual can survive contact with the university hiring file. The journal will not chase an impact factor, and it won't advertise a CiteScore. It will not promise authors a number at all. It is a publishing decision aimed at a much older problem: research assessment that still packs a thirty-five-year-old ratio into every tenure case.
The Journal of Trial and Error made a similar promise when it launched in 2021. It publishes negative results, replication failures, methodological dead ends, and papers whose main finding is that the original study did not hold up. Octopus, a publishing platform funded by UK Research and Innovation and launched in 2022, goes further: it publishes individual research outputs rather than journal issues, so journal-level metrics no longer make sense. eLife, a well-known biomedical journal, stopped making accept-or-reject decisions in 2023 and began publishing reviewed preprints; in 2024, Clarivate removed it from Web of Science, which meant no new impact factor. Peer Community In, a nonprofit platform launched in 2017, reviews preprints and recommends them without assigning a journal name or impact factor. Each case looks different, but the motivation is the same.
The impact factor itself is a simple mean. Eugene Garfield introduced it in the 1950s to help librarians decide which journals to subscribe to. A journal's 2024 impact factor counts citations in 2024 to articles published in 2022 and 2023, divided by the number of citable items in those two years. That number says nothing about any individual paper, yet it gets used to rank candidates, allocate funding, and in some countries, set salary bonuses. Review articles pull the mean upward; field-specific citation rates pull it sideways. So do editorial tricks: some journals encourage self-citation, and editors have been caught gaming the denominator.
The skew is not subtle. Fast-moving fields cite within two years. Mathematics and philosophy can take a decade. A review journal can post a factor of 20 without publishing a single new experimental result, while a genetics paper accumulates citations in months. Garfield himself warned against using the ratio to assess people. The warning did not stick, and the metric hardened into an administrative shortcut.
The discontent is old enough to have its own landmark documents. DORA was written in 2012 by a group of editors, publishers, and researchers who were frustrated that a librarian's tool had become a tenure criterion. The Leiden Manifesto added ten principles in 2015, including the instruction to read the research rather than the journal name. What makes the current wave of metrics-free journals different is that it arrives alongside institutional policy, not just editorial statement.
Here's the catch: hiring committees still want a number
Metrics-free only changes the system if hiring and promotion committees stop asking for the number. The Declaration on Research Assessment, known as DORA, has made that argument since 2012. Its signatories include universities, funders, publishers, and learned societies, and the text asks institutions to stop using journal-based metrics as a surrogate for the quality of individual research. The Coalition for Advancing Research Assessment, or CoARA, launched in 2022 and now brings together hundreds of research organisations across Europe and beyond. Both are real policy shifts. But a signed declaration does not rewrite the eleven minutes a committee member spends on each application.
Committees under time pressure reach for the shortcut. A number arrives precalculated, looks objective, and lets a panel move on. Journal-level metrics are embedded in national evaluation systems. China's Ministry of Education and Ministry of Science and Technology have already ordered universities to reduce the role of the impact factor, as AcademicJobs reported. Some hiring committees in Europe have adopted narrative CVs, but adoption is uneven. Utrecht University's TRIPLE framework asks committees to assess research quality, team science, open science, and professional performance rather than a journal name, yet few institutions have copied that fully.
Administrators face their own counterpressure. Global league tables reward citations and research income, which quietly pull universities back toward whatever metric is quickest to count. A department can sign DORA and still list journal impact factors in its annual report because the central planning office asks for them. Reforms that ignore the annual reporting form are theatre.
The practical gap lives in the rubric. Some university HR offices now write review forms that separate peer review quality from publication venue. A reviewer may be asked to score originality, methodological appropriateness, clarity of evidence, and feasibility. That sounds simple. It is not simple in practice. It requires committee members to read at least one output in full, compare it with a field's conventions, and document the reasoning well enough to survive an appeal. When forty files arrive in a week, the rubric becomes a box-ticking exercise unless someone budgets time for it.
Hype and reality are easy to separate here.
- Hype: A journal with no impact factor removes metric pressure from authors.
- Reality: CiteScore, Source Normalized Impact per Paper, Google Scholar rankings, and Eigenfactor fill the gap quickly.
- Hype: Hiring committees will read the papers.
- Reality: Reading one paper carefully takes time, and forty applications do not.
The publisher side has its own contradiction. A journal can refuse to chase the impact factor and still benefit from citation metrics on Scopus or Google Scholar. A metrics-free label can become a marketing point without changing how promotion files are read. That is why the lesson from the indexation debate applies here: appearing in a database is not the same as being judged independently. The April 2026 Web of Science delisting review put that question back on the table, as AcademicJobs covered.
The new journal's decision is not trivial. It removes a number that editors, reviewers, authors, and the administrators who read promotion files can no longer use as a proxy. But the next test is whether university administrators who sit on hiring panels notice. The eLife case showed what happens when a major journal steps outside the system: Clarivate removed it, and authors were left to explain what a reviewed preprint means to a committee that wanted a number. That is not an argument against the change. It is the price of moving first.
The metric also has a different history in different disciplines. In biomedicine, a delisting is a professional risk because grant reviewers know journal names. In the humanities, where monographs and edited volumes dominate, the journal impact factor has always been a poorer fit, which is one reason metrics-free experiments sometimes start outside the life sciences. But the risk differs by field, and a policy that works for a linguistics journal may not transfer to an oncology department.
Some research funders are trying to lower that price. UK Research and Innovation funded Octopus because the platform makes the article the object of review, not the journal. Peer Community In has recommended thousands of preprints across biology and ecology without a journal title, as its public lists show. CoARA signatories are supposed to translate their commitments into promotion guidelines. The problem is that tenure clocks do not pause while assessment culture catches up. A postdoc applying this year still faces panels that are more comfortable ranking journals than reading preprints.
University leaders who sign DORA but leave promotion criteria untouched are making a promise they cannot keep. The committee will read the application form, not the declaration. If the form still asks for journal impact factor, a candidate who lists none is not making a protest; they are making a gap in the file. The new metrics-free journal cannot fill that gap until the form changes.
Stephen Curry, a structural biologist at Imperial College London and a long-time impact factor critic, has put the case steadily: the metric was designed for librarians, not for judging scientists. It has never borne the weight institutions place on it. If a new metrics-free journal forces hiring committees to read the actual work, the launch matters. If it only removes a logo from a website, the ritual will continue.
Photo by Markus Winkler on Unsplash