Role profile: data scientist
A data scientist identifies complex business problems and works to get maximum value from data. They work as part of a multi-disciplinary team with data architects, data engineers, analysts, and others.
Data scientists work in an open, transparent and collaborative manner. They share good practice and work to improve the quality of outputs.
You can find out more about working as a data scientist in the Data Science Competency Framework.
Typical role responsibilities
Data scientists:
- work with policy and operations colleagues to understand where data science can add value
- support strategic and operational decision making
- source, access, manipulate, and engineer data processes with data – these data typically have characteristics of volume, velocity, and variety
- create credible statistical models from the data and use the best coding practices to create reproducible work
- use other technical and analytical standards from across government and industry
- work within the data science ethics framework
- are open-minded and demonstrate strong intellectual curiosity
- use techniques and knowledge from across the scientific spectrum
- explore and visualise the data to present the ‘story’ of data in a meaningful way – they communicate with a range of technical and non-technical audiences
- use a range of data analysis tools and techniques, including open source – some of these techniques must be learnt quickly, as needed
- continuously expand a range of technical skills in addition to their leadership and communications development
- encourage and support data science skills in other teams – this involves understanding the variety of functional roles that relate to data science and how they can be applied to solve business problems
Skills
There are several important skills that data scientists need to be successful in their role.
Find out more about skill level definitions.
You must be able to:
- use your programming experience in a range of coding practices to build scalable data products in a technology context
- use agile techniques and work as part of a team
- use R, Python or other programming tools to analyse data and create products
- work with developers and data scientists using appropriate tools – for example, GitHub
- identify and reuse common approaches and code
- document and test code to ensure its quality
- test data science products before they are launched
- understand the organisation’s technology stack
- get code into production by working with developers and architects
You will be expected to demonstrate these skills at different levels depending on the seniority of your role.
Trainee data scientist
As a trainee data scientist, you must have an ‘awareness’ skill level.
Associate data scientist
As an associate data scientist, you must have a ‘working’ skill level.
Data scientist
As a data scientist, you must have a ‘practitioner’ skill level.
Principal data scientist or senior data scientist
As a principal data scientist or senior data scientist, you must have an ‘expert’ skill level.
Lead data scientist
As a lead data scientist, you must have an ‘expert’ skill level.
Head of data science
As head of data science, you must have a ‘working’ skill level.
You must be able to approach new questions, opportunities, data, and techniques with an inquisitive mind.
You must work to:
- improve methods
- maximise insights
- find creative solutions to problems
You will be expected to demonstrate these skills at different levels depending on the seniority of your role.
Trainee data scientist
As a trainee data scientist, you must have an ‘awareness’ skill level.
Associate data scientist
As an associate data scientist, you must have a ‘working’ skill level.
Data scientist
As a data scientist, you must have a ‘practitioner’ skill level.
Principal data scientist or senior data scientist
As a principal data scientist or senior data scientist, you must have an ‘expert’ skill level.
Lead data scientist
As a lead data scientist, you must have an ‘expert’ skill level.
Head of data science
As head of data science, you must have a ‘practitioner’ skill level.
You must be able to apply analytical methods to reach comprehensive conclusions. These methods include:
- exploratory data analysis
- data visualisation
- statistical testing
You must be able to use analytical approaches to interpret data confidently, and summarise and describe data to support decision making.
You will be expected to demonstrate these skills at different levels depending on the seniority of your role.
Trainee data scientist
As a trainee data scientist, you must have an ‘awareness’ skill level.
Associate data scientist
As an associate data scientist, you must have a ‘working’ skill level.
Data scientist
As a data scientist, you must have a ‘practitioner’ skill level.
Principal data scientist or senior data scientist
As a principal data scientist or senior data scientist, you must have an ‘expert’ skill level.
Lead data scientist
As a lead data scientist, you must have an ‘expert’ skill level.
Head of data science
As head of data science, you must have an ‘expert’ skill level.
You must be able to understand and comply with relevant frameworks, standards and laws as they change and develop. These include the:
- General Data Protection Regulation (GDPR)
- Data Protection Act (DPA)
- Equalities Act
You must be able to work with:
- legal colleagues
- policy and public colleagues
- other government departments (OCDDO)
- ministers
You should be able to clearly communicate ethical issues and ideas to these colleagues, while storing and treating data appropriately.
You will be expected to demonstrate these skills at different levels depending on the seniority of your role.
Trainee data scientist
As a trainee data scientist, you must have an ‘awareness’ skill level.
Associate data scientist
As an associate data scientist, you must have a ‘working’ skill level.
Data scientist
As a data scientist, you must have a ‘practitioner’ skill level.
Principal data scientist or senior data scientist
As a principal data scientist or senior data scientist, you must have an ‘practitioner’ skill level.
Lead data scientist
As a lead data scientist, you must have an ‘expert’ skill level.
Head of data science
As head of data science, you must have an ‘expert’ skill level.
You must be able to:
- work with data scientists and developers to share skills
- ensure your team can access the right technologies to create products, awareness and understanding of data science techniques – this includes machine learning, natural language processing, and other techniques
- develop your own data science skills – this includes continually engaging with and updating your continuous professional development (CPD)
You will be expected to demonstrate these skills at different levels depending on the seniority of your role.
Trainee data scientist
As a trainee data scientist, you must have an ‘awareness’ skill level.
Associate data scientist
As an associate data scientist, you must have a ‘working’ skill level.
Data scientist
As a data scientist, you must have a ‘practitioner’ skill level.
Principal data scientist or senior data scientist
As a principal data scientist or senior data scientist, you must have an ‘practitioner’ skill level.
Lead data scientist
As a lead data scientist, you must have an ‘expert’ skill level.
Head of data science
As head of data science, you must have an ‘expert’ skill level.
You must be able to:
- understand the organisation’s processes, data, and strategic priorities
- apply data science techniques to produce and present data science products effectively
- design solutions that meet user needs and keep user focus, while allowing for agreed cross-government design and ethics standards
- work towards the successful creation of scalable data science products for the organisation
- establish maintenance requirements
You will be expected to demonstrate these skills at different levels depending on the seniority of your role.
Trainee data scientist
As a trainee data scientist, you must have an ‘awareness’ skill level.
Associate data scientist
As an associate data scientist, you must have a ‘working’ skill level.
Data scientist
As a data scientist, you must have a ‘working’ skill level.
Principal data scientist or senior data scientist
As a principal data scientist or senior data scientist, you must have an ‘practitioner’ skill level.
Lead data scientist
As a lead data scientist, you must have an ‘expert’ skill level.
Head of data science
As head of data science, you must have an ‘expert’ skill level.
You must be able to select and use the most appropriate tools and techniques to manipulate and transform the data for data science products. You must be able to work closely with data engineers.
You will be expected to demonstrate these skills at different levels depending on the seniority of your role.
Trainee data scientist
As a trainee data scientist, you must have an ‘awareness’ skill level.
Associate data scientist
As an associate data scientist, you must have a ‘working’ skill level.
Data scientist
As a data scientist, you must have a ‘practitioner’ skill level.
Principal data scientist or senior data scientist
As a principal data scientist or senior data scientist, you must have an ‘expert’ skill level.
Lead data scientist
As a lead data scientist, you must have an ‘expert’ skill level.
Head of data science
As head of data science, you must have a ‘practitioner’ skill level.
You must be able to use different product delivery methods and phases to help make decisions. This includes agile and waterfall techniques.
You should also be able to use your knowledge of product management to work with a range of professionals and deploy data science products into the organisation, where appropriate. You should ensure data and data science is used effectively to support products and services.
You will be expected to demonstrate these skills at different levels depending on the seniority of your role.
Trainee data scientist
As a trainee data scientist, you must have an ‘awareness’ skill level.
Associate data scientist
As an associate data scientist, you must have a ‘working’ skill level.
Data scientist
As a data scientist, you must have a ‘practitioner’ skill level.
Principal data scientist or senior data scientist
As a principal data scientist or senior data scientist, you must have a ‘practitioner’ skill level.
Lead data scientist
As a lead data scientist, you must have an ‘expert’ skill level.
Head of data science
As head of data science, you must have a ‘practitioner’ skill level.
Sample career pathway
The data scientist career path shows some of the common entry and exit points in the role. It also shows the typical skill levels needed.
You can enter a data scientist role from another analytical profession, or from other professions, such as the Digital, Data and Technology (DDaT) profession. You can also exit the role to join any of these professions.
The diagram shows a potential career path. It shows that you can enter or leave a role from a wide range of backgrounds and experience levels. For example, you could become a data scientist by developing your skills in an associate role. You could continue to move up the levels in the career path by taking on more senior data scientist roles. Or you could develop your skills by working in a technical specialist role in an analytical or digital profession. You could also develop the necessary skills by working in a profession agnostic role outside of these professions.
A role that could be done by any person with the relevant skills or experience from any profession.
This could be a ‘badged’ or professional role that is subject to entry requirements and development.
Beyond the head of data science role, you could go into more senior leadership roles. These roles require broader analytical understanding, and the ability to lead multi-disciplinary teams.