How to hire a data professional
By Francis Adio, Business Analyst; and Lydia Monnington, Data Science Lead and long-term DataKind UK volunteer
Our guidance is here to help you find the right data person for your organisation. This is the second in a short series about the who, what, when, why, and how of hiring a data professional, so before you dive in, take a look at the previous resource, When to hire a data professional.
Ready to recruit?
Common challenges when hiring data professionals include vague job descriptions and unrealistic expectations, where one hire is expected to do it all, from designing infrastructure to creating dashboards. This can lead to high turnover, an expense we all want to avoid.
The work you've done to identify your organisational goals and the relevant skills you want (see our handy previous article!) should give you a good idea of your immediate and long term needs, and how you will split that work between roles.
Drafting your job spec
The goal of your job spec is to give candidates a clear vision of the role, including what work they will do, and the skills needed for success, to draw in the people you are looking for. Using the outcomes and skills you’ve identified, write a list of the key skills and nice-to-have skills. Be specific and define your current needs and deliverables. Use plain English, short sections, and action-focused headings. And make sure you don’t combine three jobs into one!
Try to keep the key skills list short: 2-3 technical skills and 2-3 non-technical skills, plus your general requirements for working with your organisation and its mission. Make it clear which are essentials and which are nice to have, and add some colour about why these are important to you. Be realistic and avoid superhuman specs - you don’t need Python and Machine Learning to use Excel and Power BI.
Be inspired by job specs from similar organisations, what you find compelling, and why. And always follow best practice for an inclusive hiring approach - people are put off when lists of requirements are too long!
Selecting salaries
Data roles in third sector organisations can be paid less than their commercial equivalents. This makes recruitment competitive and retention difficult, especially if progression pathways are limited. Be explicit about any benefits the role includes, especially investment and support with development and training.
After assessing your existing needs and budget, it’s worth reviewing market data on salaries, such as research from Data Orchard on Charity Data salaries. Benchmark the salaries for similar roles with other organisations by searching for them on other job boards or using salary benchmarking websites.
Hiring vs consultancy
If you aren’t sure you need a permanent role added to your team, you will still need someone in-house with the expertise to manage and monitor any data work. Consultants or outsourced support are most effective for time-bound initiatives or highly technical projects that exceed your current team’s capacity. In-house hires are best suited to ongoing needs, especially those tied to compliance, governance, and recurring, strategic reporting.
It’s essential to maintain internal oversight whether you are outsourcing your needs or not, through a team member who can safeguard data security, ensure continuity, and uphold accountability to funders and stakeholders.
| Role | Benefits | Limitations | Recommendation |
|---|---|---|---|
| Business Analyst | Independent perspective on workflows and processes | Limited knowledge of organisational culture and day-to-day operations | Keep in-house when a data manager or analyst cannot be hired. Consultants can advise, but internal staff should lead |
| Data Manager | Avoids hiring costs | Loses governance, GDPR oversight, and accountability to funders/trustees | Essential for strategy, compliance, and coordination. Keep in-house - can be part time or fractional |
| Data Analyst | Advanced skills for funder reports or project evaluation | External analysts may misinterpret your programmes’ context or deliver generic outputs | Keep in-house when a data manager cannot be hired. Outsource only for specialist one-off analysis, with oversight |
| Data Governance / Quality | Consultants can help set up GDPR compliance quickly | Long-term accountability cannot be outsourced; risks to donor/beneficiary trust | Outsource/volunteer. Accountability can be achieved with internal oversight |
| Database Administrator | Cost-effective for routine maintenance, upgrades, or migrations | Reduced control over sensitive donor/beneficiary data; dependency on providers | Outsource/volunteer. Access control and project oversight must be internally managed |
| Data Engineer | One-off integrations (e.g. CRM → reporting dashboards) without hiring full-time staff | Poor documentation - systems may fail once consultants leave | Outsource/volunteer to build projects, but ensure internal oversight |
| Data Entry / Processing | Volunteers or low-cost outsourcing reduces staff burden on repetitive tasks | Risk of errors, inconsistencies, and poor data quality if unchecked | Outsource/volunteer – but only with strong in-house quality checks |
Posting your role
When posting your role, it’s fine to use whatever boards or agencies you might usually use, such as CharityJob, Third Sector, or the Guardian. Publicise and share the role as you would any other, to your donors, volunteers, supporters. Do your current staff, volunteers, or beneficiaries know anyone who might be great? Are there any staff members who have been doing a lot of data work and would like an opportunity to do more?
To find a wider audience, social networks allow you to tag and share with professional groups - tags like ‘DataAnalystJob’ can help it appear in front of a lot more eyes. Patrons, sponsors, directors, and trustees may have interesting networks and connections they can pass it on to. Share with relevant communities and groups such as Black in Data, The Data Lab, or PyLadies.
DataKind UK also runs a jobs board that aims to bridge the data/third sector skills gap!
Selection process
When reviewing applications, look for examples of each of your key skills. A key differentiator for data roles, in particular business-facing roles like data analyst or data scientist, is whether they talk about the result of the work they have done on the wider business. If their CV is well-structured and compelling, with good examples of working with stakeholders, this points to decent communication.
Build a consistent interview process
To hire fairly, it's important to have a consistent process to assess candidates. Data professionals frequently follow non-linear career paths, because they often come from varied educational and career backgrounds. As CVs alone may not fully indicate their suitability, the process should include both practical and skills-based elements.
Map each skill you are looking for, and ensure you have a marking rubric for each. This could look like a grid of each of your key skills, and what a No Hire, Borderline Hire and Strong Hire looks like for each.
Are they meeting the bar you expect on knowledge or skills?
Are their answers clear and understandable by non-technical people?
Are their answers well-structured and not too long?
Examples for a Data Analyst in a medium-sized team
| Type | No Hire | Borderline Hire | Strong Hire | |
|---|---|---|---|---|
| Skill-based These are your classic “Tell me about a time when…” questions. Ask them to talk about relevant experiences and concrete examples, looking for skills such as problem-solving, communication, collaboration, and adaptability. | Weak or no examples given. | Good but not great examples. | Compelling examples strongly related to the question asked. | |
| Domain knowledge Ask them to explain key concepts in a relevant skill area, such as statistics, and ask follow-up questions. | Not familiar with the concept, or struggles to communicate it. | Some examples of concepts and familiarity with a few uses. | Explains concepts in business terms and gives clear, practical examples of how they have applied them. | |
| Practical test Identify a task similar to something someone in this role would actually need to do, e.g. ask them to analyse some of your data, and share their key insights or visualisations. Simplify it so it can be done in about 30 mins by an experienced member of the team. Give the candidate 60 minutes to complete the task and then check in with them to share their results. | Few insights found. Visualisations unclear. | Most insights found. Some thought on how to communicate key points. Good discussion on options. Acceptable visualisations chosen. | Found all the insights we expected from the data, plus a few extra. Able to explain the insights and their process clearly. Selected suitable visualisations for the task. Effective use of colour or annotation to draw attention to key points. | |
| Coding test For technical tests that you are less familiar with, there are often tools available online. Design a question related to the languages or SQL skills you need. Ask them to write some code to solve it. | Struggled with the question. Didn’t respond well to hints or feedback. | Got through >80% of the question. Needed hints to solve problems but used them effectively to get to the answer. | Able to complete the question quickly with no errors. | |
Conclusion
Hiring data professionals differs from other recruitment because the roles are diverse and highly specialised. Success depends on:
Starting with your organisational needs.
Matching the right role to the right requirement.
Hiring at the right time to meet organisational priorities.
Writing realistic job descriptions.
Selecting for problem-solving, communication, and adaptability.
By approaching data hiring carefully, you can avoid common pitfalls such as mismatched roles or unrealistic expectations. The right hire ensures not only effective data management but also evidence-based decision-making across the organisation, and a well-informed team.
AI transparency statement: AI was used for some grammatical corrections.
