Data Competency Framework: Structure, Scope and Named Examples
Data roles are among the most inconsistently defined in any organisation. A data analyst in one business does something categorically different from a data analyst in another, and the range of competency expectations attached to data engineering, data science, or data governance roles varies even more. A data competency framework is the instrument organisations use to define what data professionals are expected to be able to do, at what level, and across what domains. Most of the ones I see are either too technical (a list of tools and methods) or too behavioural (a set of generic competencies lifted from a leadership framework). The ones that work are neither.
What Is a Data Competency Framework
A data competency framework defines the competencies required across data roles in an organisation, expressed at proficiency levels that distinguish between foundational, intermediate, and advanced application. It covers the full range of data work: from data literacy and data governance through to data engineering, analytics, modelling, and data leadership.
A well-designed data competency framework does two things that a generic competency framework or a skills list cannot. It reflects the specific domains of data work, which differ from general digital or technology roles, and it defines competency at the level of specificity needed to make workforce decisions about data professionals rather than just digital professionals broadly.
Why a Data Competency Framework Exists
Data functions have expanded rapidly, and the vocabulary used to describe data competency has not kept pace. Organisations routinely conflate data literacy (the ability to read and work with data as part of any role) with data analytics competency (the ability to perform analytical work on data), and both with data engineering or data science competency (the ability to build data infrastructure or develop predictive models). A framework that does not distinguish between these domains cannot support meaningful hiring, development, or succession decisions in a data function.
The second driver is the genuine diversity of data roles. A data analytics competency framework addresses one slice of the data landscape. A broader data competency framework needs to account for the full spectrum: data governance, data management, data architecture, analytics, science, and the leadership and communication competencies that sit across all of them.
Research on data science competency in organisations identifies that data competency is not a single construct. Effective data roles require the integration of technical, analytical, and domain-specific proficiencies, and no single individual can be expected to hold all of them at an advanced level. A framework that treats data capability as a unified dimension rather than a set of distinct domains produces assessments and hiring decisions that systematically misread what the work requires.
Named Data Competency Frameworks
APS Data Capability Framework
The Australian Public Service Commission's Data Capability Framework is one of the most developed whole-of-government data competency frameworks available publicly. Version 2, released in 2025, outlines 26 data-specific capability areas associated with working with data across the Australian Public Service. Each capability area has indicators covering skills, knowledge, and behaviours across three proficiency levels: foundation, intermediate, and advanced.
The APS framework is also notable for its alignment to SFIA (the Skills Framework for the Information Age), which enables greater comparability with industry and international standards. This layered approach, using SFIA as an anchor for skill definitions while building domain-specific capability indicators on top of it, is an increasingly common design pattern in public sector data frameworks.
SFIA Data Skills
The SFIA framework defines a set of data and analytics skills across its six categories, including data management, data analysis, data science, and business intelligence. SFIA skills are defined at the levels of responsibility at which they are realistically performed, from foundational application through to strategic leadership. The SFIA Foundation's work on developing data skills in the public sector demonstrates how SFIA skills can be used to build role profiles and competency frameworks for data professionals across diverse organisational contexts.
SFIA is a skills framework, not a competency framework. The distinction matters because SFIA defines what data professionals need to be able to do, described at levels of responsibility, while a competency framework defines how they are expected to work and perform. Used together, they complement each other in a way that neither provides alone.
Edison Data Science Framework
The Edison Data Science Framework, developed through European research and industry collaboration, defines competencies for data science roles across five primary areas: data analytics, data engineering, data management, research methods, and domain applications. It is used as a reference for curriculum design, professional certification, and data role definition. While more academically oriented than the APS framework, it reflects a similar commitment to distinguishing between the technical, analytical, and operational domains of data work.
How a Data Competency Framework Works in Practice
A functioning data competency framework is built around the actual structure of data work in the organisation, not around a generic idea of what "data skills" means. This typically requires starting from the role types that exist or are planned, working out the competency domains those roles draw on, and then defining what good performance looks like at each level within each domain.
The competency domains in most data competency frameworks include: data governance and stewardship, data management and infrastructure, data analytics and visualisation, data science and modelling, and data communication and storytelling. Each domain requires different skills and different proficiency descriptors. Combining them into a single undifferentiated framework loses the precision that makes the framework useful.
What a Data Competency Framework Is Not
It is not a competency framework applied generically to data professionals. General competency frameworks define behavioural expectations for a workforce broadly. A data competency framework must go further, defining the domain-specific competencies that distinguish data roles from adjacent roles. Generic leadership competencies applied to data professionals will not tell you whether someone can design a data pipeline, evaluate a machine learning model, or govern a data asset.
It is not a skills list. A list of tools or technologies, such as Python, Tableau, SQL, or Azure, is not a competency framework. It is a skills inventory. The capability framework design principles that distinguish a framework from a list, including defined proficiency levels, behavioural indicators, and connections to role-level decisions, apply as much to data roles as to any other domain.
It is not a data literacy programme. Data literacy refers to the general ability of any employee to work with data as part of their role, reading reports, understanding metrics, and making data-informed decisions. A data competency framework is for people whose primary work involves data: professionals who build, manage, analyse, or govern data systems and products. The two constructs address different populations and different questions.
Common Failure Modes
The most common failure in data competency frameworks is building a framework around tool proficiency rather than competency domains. A framework that assesses whether data professionals can use specific platforms is useful for procurement and onboarding but not for development planning or performance assessment. The tools change; the underlying competencies do not at the same rate.
A second failure is applying a single proficiency scale across all data domains. An advanced analyst and an advanced data engineer are both described as "advanced" in their respective domains, but they are doing fundamentally different work at fundamentally different levels of technical complexity. A framework that conflates these on a single scale produces assessments that cannot be compared meaningfully.
A third failure is designing the framework in isolation from the data leadership. A data competency framework that is not connected to how data roles are recruited, how performance is assessed, and how career pathways are structured becomes a document rather than a system.
Trade-offs and Constraints
A data competency framework that covers the full range of data competency will be more accurate for the diversity of data roles in a large organisation but harder to maintain and apply consistently. A narrower framework, focused on a specific data discipline such as analytics or data engineering, will be more precise but less useful for organisations that need a common language across data role families.
Frameworks that align to international standards like SFIA benefit from external legitimacy and interoperability with other organisations and credential providers. Frameworks built entirely internally may be more specific to the organisation's context but require more design effort and have no external comparator for benchmarking.
Frequently Asked Questions
What is a data competency framework?
A data competency framework defines the capabilities required of data professionals across the relevant domains of data work, expressed at proficiency levels. It covers areas such as data governance, data management, analytics, data science, and data communication, and is used for recruitment, performance assessment, and workforce planning in data functions.
What is the APS Data Capability Framework?
The Australian Public Service Commission's Data Capability Framework defines 26 data-specific capability areas for APS employees, with indicators covering skills, knowledge, and behaviours at foundation, intermediate, and advanced proficiency levels. Version 2, released in 2025, aligns to SFIA 9 to enable greater comparability with industry and international data capability standards.
Is SFIA a data competency framework?
SFIA is a skills framework for digital and technology professionals, not a competency framework. It includes data and analytics skills defined at levels of responsibility, which can be used to build data role profiles and data competency frameworks. It is a useful reference for data competency framework design but is not itself a complete data competency framework.
How many domains should a data competency framework cover?
Most comprehensive data competency frameworks cover five to seven domains: data governance, data management, data analytics and visualisation, data science and modelling, data engineering, data literacy, and data leadership or communication. The right scope depends on the range of data roles in the organisation.
How does a data competency framework differ from a data skills list?
A skills list catalogues specific technical abilities, such as tool proficiencies or method knowledge. A competency framework defines competencies at proficiency levels, with indicators describing what effective performance looks like in practice at each level. A skills list tells you what someone knows how to do; a competency framework tells you how well they do it and what the next level of development looks like.
Can a data competency framework apply to non-data roles?
A data competency framework is designed for people whose primary work involves data. Data literacy frameworks, which define the baseline data understanding expected of all employees, are more appropriate for non-data roles. The two serve different populations and should not be conflated.