SFIA View: Big Data/Data Science view

Big Data/Data Science view

Big data is data that contains a greater variety arriving in increasing volumes and with ever-higher velocity. Data Science is the capability to analyse and get value from the data through the application of a variety of methods such as mathematics, statistics, computer science, machine learning and data visualisation.

Click on image to see full-size interactive pdf with hyperlinked skill names.

The Big Data / Data Science view of SFIA has been developed to provide a quick start identification of the SFIA skills which are most relevant/illustrative for organizations to adopt and operate big data, data analytics and data science working practices.

  • This view of SFIA identifies around 40 to 50 professional skills related to data within the complete SFIA  framework of more than 100 skills.
  • The content of the SFIA framework is reviewed on a regular basis, and with the latest version of SFIA, the content has been reviewed and updated to capture the professional skills needed - whether for leading, managing and executing the adoption of effective data science practices
  • This includes the skills needed for developing a supportive culture and of new organizational capabilities for data management

As with all applications of SFIA; this view should be considered against your specific organizational context and business objectives.

  • The intention is not to draw a hard boundary around these skills or to imply that other SFIA skills aren't appropriate. In practice, once you have familiarised yourself with this view; it is likely that you will refer to the full SFIA framework for additional and complementary skill definitions.
  • The full SFIA reference guide and a spreadsheet version of all skill descriptions are available to download. You will need to be registered as a user on the site first but it is a very simple process.
  • This website provides advice and guidance on the adoption of SFIA.
  • There is also an active global ecosystem of SFIA Partners, SFIA Consultants and Practitioners. They are available for advice on adopting SFIA. Full details are available here.
  • If you represent a professional body or a framework owner and would like to collaborate with the SFIA Foundation on the development of additional SFIA views; please contact the 

For ease of use, the Big Data/Data Science skills view has been split into 4 focus areas based on different aspects of data management practices.

  • The groupings are a navigational aid only 
  • Once you are familiar with the content of SFIA they may not be needed
  • Each SFIA skill is present in only one focus area, but in practice, some skills could easily be represented in more than one focus area
  • SFIA skills for individual job roles should be selected appropriately from any of the focus areas and/or from the wider SFIA framework

Data governance

  • skills needed to align data management practices to organizational purpose through the development and implementation of effective strategies, policies and procedures 

Data culture change and organizational capability development

  • skills needed to assess the maturity and effectiveness of data management practices and develop improved organizational capabilities (people, processes and technology)

Data lifecycle management

  • skills needed to support the full data lifecycle from generation capture, maintenance, active use, publication, archiving, purging

Data foundations

  • skills needed for common foundational practices underpinning the other data management capabilities

Extract of SFIA design principles

SFIA defines levels of responsibility and skills. 
SFIA does not define jobs, roles, people, processes or general areas of activity, however important they are.
SFIA defines the essence of skills. 
SFIA is descriptive, not prescriptive. It does not define low-level tasks nor deliverables.
SFIA provides an integrated view of competency. 
SFIA recognises levels of responsibility, professional skills, behaviors or attributes, knowledge and qualifications and certifications. It shows how these fit together and how they complement each other.
SFIA is independent of technology and approach. 
SFIA does not define technology, methods, approaches or technical knowledge – these change rapidly but the underlying skills are more persistent. So, for instance, Cloud, DevOps, Agile, Big Data and digital roles etc. can be described using a combination of the SFIA skills.
SFIA does not assume or recommend specific organization structures, job or role designs. 
The SFIA skills and levels can be configured flexibly to support all organizational types and structures. It works for individuals, small and large teams, whole departments or entire organizations with thousands of employees.

Data management jobs and skills profiles

Job titles are typically used as a shorthand reference to Data Management jobs. So in organization structures and job ads we see a variety of titles. e.g. 

In addition; many other positions have data management responsibilities and skills but only as part of their overall job. e.g. 
  • Software Developer
  • Application Architect
  • Enterprise Architect
  • DevOps Engineer

We should also consider the variety of prefixes commonly used to support job grading or career pathing. e.g.

  • Head of Data Science
  • Lead Data Scientist
  • Principal Data Scientist
  • Senior Data Scientist
  • Data Scientist
  • Junior Data Scientist
  • Trainee Data Scientist

However, as in most professional disciplines, there is no common definition of what these jobs actually do. So although in common usage, referring only to a job title in isolation can be confusing and does not help organizations recruit, develop, deploy, manage and retain their valuable talent. 

Using SFIA to improve Big Data/Data Science job descriptions.

SFIA provides a common language of skills and skill levels. It is very flexible and this enables organizations to design their own team structures, roles and job titles. They can then select the appropriate configuration of SFIA skills and competency levels to match. 

  • The SFIA skills for roles responsible for big data and data science should be selected based on an analysis of the role's accountabilities and responsibilities. 
  • To provide the necessary focus, aim for no more than 6 to 8 SFIA skills per role (less if possible) 
  • The required skills can be selected from the full range of skills in SFIA 
  • The skill levels chosen should be based on the responsibility levels of the role and aligned to SFIA's generic attributes for levels of responsibility

Focussing on job responsibilities, and the SFIA-defined professional skills and responsibility levels provide a much clearer definition of the requirements of the job. This, in turn, supports people management related activities such as recruitment, skills assessment, professional development, and performance management.

Standard or model skills profiles for Big Data/Data Science roles

The SFIA Foundation is looking to develop and publish a set of standard skills profiles for some of the common industry roles. This is work in progress and if you would like to know more and/or contribute please contact the .

Where is the Data Science skill in SFIA?

Following these principles, it should be clear why there is no single "Data Science" skill in the SFIA framework, There are many SFIA skills and attributes at multiple levels which organizations can align to their Data Scientist roles. 

For example;

Potential Data Scientist Skills in SFIA

.. and their Generic Attributes

Analytics

Data visualisation

Programming/software development 

Research

Innovation

Data modelling and design

Data management 

Business modelling

Emerging technology monitoring

Measurement

Methods and tools

Autonomy

Influence

Complexity

Knowledge

Business Skills

  • Communication
  • Personal Work Scheduling
  • Teamwork or otherwise
  • Problem Solving
  • Ethics, Code of Conduct
  • Security