Research Data Steward
A research data steward helps researchers look after the data behind their work - making sure it is organised, documented, stored securely, legally handled and, where appropriate, shared openly so others can reuse it. They typically sit in a university library, research office or IT service, or in a research institute, NHS trust, government body or research-intensive company, and spend their time advising researchers, checking datasets and metadata, and running the systems and policies that keep research data findable and reusable for years after a project ends.
- Starting salary
- £26k–£34k
- Live graduate jobs
- Search all jobs
- Main way in
- Direct entry into an assistant or junior data steward /…
- They usually ask for
- A degree is normally expected
Approximate graduate salary
Entry-level posts typically start somewhere around £26,000–£34,000, though this varies a lot - university posts sit on published pay scales and tend to be at the lower end outside London, while data steward roles in industry, pharma or government data governance teams can start noticeably higher. Treat these as rough approximations only.
1What you'd actually do
- Answering queries from researchers and PhD students - things like 'can I put this dataset on a public repository?', 'where do I store interview recordings containing personal data?', 'how do I write a data management plan for this grant application?'
- Reviewing data management plans (the short document funders require explaining how data will be collected, stored, shared and preserved) and suggesting fixes before a grant bid goes out
- Checking datasets that researchers want to deposit in a repository: is the file format one that will still open in ten years, is the README file clear, are the variable names explained, is there a licence attached, does it need a DOI (a permanent web identifier so the dataset can be cited)
- Writing or improving metadata - the structured description of what a dataset is, who made it, when, and under what conditions it can be reused - and checking it meets the relevant standard for the discipline
Show 3 moreShow fewer
- Running training sessions and drop-in clinics for research staff on data management, open data requirements and handling sensitive data under UK GDPR
- Working with IT, information governance and research ethics colleagues on storage, access controls and anonymisation for sensitive data, especially health or personal data
- Doing some administration of the systems themselves - the institutional data repository, catalogue entries, storage quotas, chasing researchers whose projects have closed but whose data has not been archived
2How graduates get in
- Direct entry into an assistant or junior data steward / research data officer post, often advertised by university libraries or research services. This is the most common route and vacancies are irregular rather than annual.
- Via a related university role first - repository assistant, library assistant, research support administrator, scholarly communications assistant - then moving sideways as a data-focused post opens. Very common, because institutions often prefer to recruit someone who already understands how research works internally.
- From a research background: finishing a PhD or a research assistant contract and moving into data support. Also common, particularly for stewards embedded in a specific discipline (bioinformatics, social science data, clinical trials) where subject knowledge is needed.
- Through a library and information science master's (MA/MSc Librarianship, Information Management, Digital Curation), sometimes studied part-time while working. Not compulsory, but it is a recognised route and some posts ask for it or expect you to work towards it.
Show 2 moreShow fewer
- Graduate trainee schemes in academic libraries or in research institute data/informatics teams. These exist but are limited in number, and not all of them include research data work.
- Occasionally through data-focused graduate routes in government, NHS or public research bodies, where the job may be titled data steward, data manager or information governance officer rather than research data steward.
3What employers ask for
- A degree is normally expected, and the subject matters less than the mix of skills - science, social science and humanities backgrounds all appear. For discipline-specific posts (genomics, clinical research, environmental data) a relevant subject degree is often genuinely required.
- A 2:1 is a common expectation but this is softer than in graduate schemes; relevant experience frequently outweighs classification, and some posts are advertised on experience alone.
- Experience of research - your own dissertation, a research assistant job, a lab placement or a PhD - because credibility with researchers depends on understanding how projects actually run. Some employers will accept strong library or data admin experience instead.
- Working knowledge of UK GDPR and data protection as applied to research, plus research ethics basics. Employers usually expect awareness rather than formal qualification at entry, and will train you up.
Show 2 moreShow fewer
- Practical familiarity with handling data files - spreadsheets, structured formats such as CSV, XML or JSON, and often some scripting in R or Python. The depth required varies hugely: some posts are almost entirely advisory, others expect you to write code to validate or convert datasets.
- A postgraduate qualification in librarianship, information management or digital curation is an asset and sometimes required for more senior posts; CILIP (the professional body for library and information professionals) chartership is valued in university settings but rarely required to start.
4Skills that matter
Metadata and documentation discipline
Most of the job is making sure someone who has never met the researcher can understand and reuse a dataset years later, which comes down to consistent, standards-based description.
Translating policy into practical advice
Funder mandates, UK GDPR and institutional policy are written in dense language, and researchers need a plain answer about what they can and cannot do with their files.
Data wrangling with spreadsheets and basic scripting
You will regularly open messy files to check structure, spot personal data that should not be there, convert formats or run validation before a dataset is published.
Diplomacy with busy, senior academics
You often have to tell a professor their dataset cannot be shared as it stands, with no authority over them, so persuasion matters more than process.
Understanding of research workflows in at least one discipline
Advice that ignores how a lab, a cohort study or a qualitative project actually generates data gets ignored, so knowing the practicalities buys you credibility.
Training and explaining to mixed audiences
A large part of the role is delivering short sessions and one-to-one clinics to people with wildly different technical confidence.
Putting these on a CV? We counted which skills 12,914 live graduate adverts ask for, and what to do with the answer — part of our application advice.
5Where it leads
Assistant or officer level: handling the day-to-day query queue, repository deposits and metadata checks under supervision. Entry posts are often fixed-term, tied to a project or grant.
Research data steward / research data manager: owning an area of service, reviewing data management plans independently, running training programmes and advising on sensitive data. Moving up usually takes a few years but timelines vary widely because progression depends on posts becoming vacant rather than on a fixed ladder.
Senior or lead roles: head of research data management, scholarly communications manager, or research data service lead, with responsibility for institutional policy, budgets and a small team.
Sideways moves are at least as common as upward ones - into research information governance, data protection (some stewards go on to qualify as data protection practitioners), repository or systems management, research funding and policy, or informatics roles in the NHS or a research institute.
Outside academia, the skills transfer into data governance, data cataloguing and data quality roles in government, pharma, consultancies and large companies, where equivalent jobs are usually titled data steward or data governance analyst and often pay more.
What people get wrong
“It is a library job about filing documents.”
Much of the work is technical and legal - file formats, identifiers, access controls, anonymisation, licensing and funder compliance - and in some posts you will spend more time with IT and information governance colleagues than with librarians.
“You need to be a programmer or data scientist.”
Some posts want scripting, but many are advisory and policy-facing; the core competence is understanding research practice and data standards, and plenty of stewards come from humanities and social science backgrounds.
“You analyse the research data yourself.”
You almost never do the analysis - your job is to make sure the data is well described, lawfully handled and preserved so that the researchers, and later other people, can analyse it.
“It is a quiet back-office role.”
It is heavily people-facing and often involves negotiating with researchers who see data management as bureaucracy, under deadline pressure from grant submissions and publication requirements.
6Where this varies
The job title covers quite different jobs. In a university, it is usually a central service role split between policy advice, training and repository work, often on a fixed-term contract funded by a project or grant. In a research institute or large clinical/genomics programme, a data steward is often embedded in one research group and much more technical, working with controlled-access data, data standards and pipelines. In government, the NHS or industry, 'data steward' frequently means something closer to data governance - owning definitions, data quality rules and catalogue entries for business data rather than research outputs - and the research element may be minimal. Read the advert carefully: the balance between advisory, technical and governance work differs enormously, and so does the pay.
General guidance about the role across the UK market, not about any specific employer. Entry routes and requirements vary — always check the individual job advert.