Ask senior researchers what they are proudest of and many will not name a paper. They will name a person — a doctoral student who became an independent scientist, a postdoc they steered through a difficult first grant, a junior colleague they helped find their feet. Supervising students and mentoring early-career researchers is among the most consequential work in academic life, with effects that ripple across whole careers. And yet, for a long time, it has been almost invisible in the formal record: it does not appear in a citation count, it rarely earns a byline, and it can be hard to point to on a CV. This article looks at how that is beginning to change, drawing on the contribution definitions in the mentorship and career stages domain of the CASRAI Dictionary.
Why the work has been invisible
The invisibility is structural, not accidental. Traditional research assessment is built around countable outputs — publications, citations, grant income — and mentorship produces none of these directly. Its products are people and their subsequent achievements, which are attributed to them, not to their mentor. The mentor’s contribution is real but diffuse, exercised over years through conversations, feedback, advocacy and example. None of that fits neatly into a metric, and what does not fit a metric tends to drop out of formal recognition.
The cost of this invisibility is not only unfairness to good mentors. It is an incentive problem: if mentorship does not count, the rational career move is to spend less time on it, which damages exactly the developmental relationships that sustain the research workforce. Making the work visible is therefore not a courtesy — it is a way of keeping it valued and done well.
The CRediT Supervision role
The first piece of structure comes from contributorship. The Contributor Roles Taxonomy includes a Supervision role, defined as oversight and leadership responsibility for the research activity planning and execution, including mentorship external to the core team. Its inclusion is significant: it acknowledges, in the same controlled vocabulary used for writing, analysis and data curation, that guiding and overseeing the work is itself a distinct contribution worth naming.
The Supervision role lets a contribution statement record who provided that leadership and mentorship on a given piece of work, rather than leaving it to be inferred from author order or omitted entirely. It is not a complete solution — a single role on a single paper does not capture a multi-year supervisory relationship — but it establishes the principle that supervision is a recognised category of contribution, and it makes that recognition machine-readable wherever CRediT is applied. The full set of roles, including Supervision, is set out in our overview of the CRediT roles.
What the narrative CV adds
Where CRediT captures supervision on a specific output, the narrative CV captures the longer arc. A narrative CV asks researchers to describe their contributions in prose — under headings that typically include the development of individuals and the wider research community — rather than presenting an undifferentiated list of publications. This format is precisely suited to mentorship, because mentorship is a story, not a number: it is the account of who you helped, how, and what they went on to do.
Under a narrative CV, a researcher can describe the students they have supervised to completion, the early-career colleagues they have supported, the training they have built, and the ways they have contributed to a healthy research culture — and have all of it sit as legitimate evidence of contribution rather than as soft, unscored extra. Several major funders have adopted narrative formats with explicit sections for exactly this kind of work, signalling that developing people is part of what they are assessing. Our guidance on the narrative CV for authors sets out how to write these sections so they evidence contribution rather than merely assert it.
Evidencing mentorship honestly
As with any contribution, the value of recording mentorship depends on doing it honestly. Good evidence is specific and, where possible, corroborable: named individuals supervised and their outcomes, training programmes designed and delivered, roles in formal mentoring schemes, and changes to research culture that others can confirm. The aim is not to inflate — claiming credit for the achievements of people you barely advised is its own kind of misconduct — but to make visible the real developmental work that has genuinely shaped careers. A narrative that describes a mentee’s trajectory and the mentor’s specific part in it is far more credible, and far more useful to an assessor, than a vague claim to be ‘a dedicated mentor’.
Recognition across career stages
Mentorship also looks different at different career stages, and good recognition accounts for that. An early-career researcher may mentor undergraduates and contribute to peer support; a mid-career researcher may supervise doctoral students and lead a group; a senior researcher may mentor across an institution and shape its developmental culture. Treating these as a single undifferentiated category undervalues both the junior researcher whose mentoring is real but modest in scope and the senior one whose influence is institution-wide. Recognising mentorship well means recognising it appropriately to the stage.
The broader shift here is part of the move toward assessing the whole researcher — valuing the development of people and culture alongside the production of papers. Structured contribution roles such as Supervision, narrative CV formats that make space for developmental work, and a consistent vocabulary for describing it all play their part. That vocabulary is maintained in the CASRAI Dictionary, helping ensure that when a researcher says they have mentored, supervised or developed others, the claim can be recorded, understood and credited across the systems that track research careers.