Conclusion: Regardless of its digital strategy, many organisations have not been positioned to properly leverage the digital and data assets that are available to them. A Chief Data Officer (CDO) role can improve this situation by advancing an organisation’s data portfolio, curating and making appropriate data visible and actionable.

The CDO position is appropriate for all larger organisations, and small-to-large organisations focused on data-driven decision-making and innovation. These organisations benefit from a point person overseeing data management, data quality, and data strategy. CDOs are also responsible for developing a culture that supports data analytics and business intelligence, and the process of drawing valuable insights from data. In summary, they are responsible for improving data literacy within the organisation.

Observations:

The evolving role of the CDO: Organisations that are looking to hire a CDO need to recognise that this role is evolving from a defensive strategy to an offensive strategy:

CDO’s Defensive Strategy: In this strategy, the CDO helps the organisation respond to potential threats based on historical data. As detailed in ‘Workforce Transformation: The Four Operating Models of Business Intelligence’, in a defensive strategy, the CDO’s primarily focus on the first two operating models: traditional business reporting (i. e. looking at the logbook), and data visualisation (i. e. looking in the rearview mirror).

CDO’s Offensive Strategy: In this strategy, the CDO helps the organisation leverage data to predict and shape business direction. This strategy encompasses the two remaining operating models of business intelligence: key performance indicators (i. e. speedometer/dashboard), and decision support (i. e. looking at the GPS).

To be successful with implementing an offensive strategy, CDOs must now act as change-agents (maturing data literacy within the organisation). They also need to take ownership of delivering value from data, measuring and communicating the business impact of any data projects.

The new CDO agenda

1. Analytics prioritisation

Analytics prioritisation is a principle of knowing what is the most important data or analytics the organisation needs. It also requires a CDO to identify the driving points for the analytics prioritisation to work.

This is possible using the concept of Value Engineering, wherein CDOs implement enterprise-wide analytics priorities to meet the demands from the business. CDOs need to start with helping management to understand how to drive business value from data and analytics. This is done by first meeting with the organisation’s executives to define the ’data and analytics user needs’.

Once the high-level data and analytics user needs are defined, the CDO should consider how they will transition the organisation from a compliance and reconciliation mindset to focusing on analytical value and knowing the data cost. This requires evaluating the current digital literacy of the organisation, and then building a strategy to improve that maturity to a level that is appropriate to meet the data and analytics needs.

2. Business initiatives

Business initiatives should be a driving force for CDOs. CDOs focus on the importance and value of data to support business initiatives and how to locate and harness the right data at the right time. For example, to harness the correct analytics, data should be retrieved at the right time and correct location (data catalogue, data warehouses, data lakes, etc.).

CDOs’ involvement with business initiatives is a combination of technology and psychology. They need to balance the expectations of stakeholders with the reality of the data available, for the form in which the data will be delivered (how it will be visualised or be used to make recommendations and predictions). The CDOs are also responsible for business stakeholders being comfortable with the data: i. e. the data quality must be appropriate to the business need, not necessarily being ‘perfect’.

3. DataOps centred model

CDOs will champion the development of a DataOps. Data analytics are increasingly getting more complex and require a DataOps model, which is made up of three key elements:

• Discover: Organisations can quickly leverage data to produce new insights that promote time to value and highlights data and operation challenges. There are two approaches to achieving agile discovery: data literacy and data democratisation.
  • Data literacy: The ability to read, write, and communicate data within an organisation.
  • Data democratisation: The ability to make digital information accessible to both technical and non-technical users of information systems without requiring the involvement of IT.
• Operationalise: Deliver a data catalogue, data pipeline, and operational analytics to drive demonstrable business impact in a repeatable manner. This is possible using a data integration platform combined with DataOps methodology.
• Measure: Performance of operation and measure the business impact of deployments.

4. Champion data ownership

Data ownership is when business stakeholders take ownership and responsibility for data – and importantly data quality-related issues. Data ownership is domain-oriented. The two key concepts around business domains ownership are:

• Source-oriented domains: Business stakeholders know where to find data and are knowledgeable about how business transactions happen. For instance, it is important that there is someone who is responsible for the application design. There is a point person to explicitly define a term and tag the data with its intended purpose. Too often, data users encounter two systems that use different terms pertaining to one particular data item.
• Consumer-oriented domains: Every data and its parameters are available and searchable in the catalogue. This addresses the issue that despite different silos, everybody is synchronised and will not interpret varied results and outcomes.

To implement a data strategy is a discipline: one towards putting perspective into ownership of the data. Furthermore, here is a sample framework on how to understand the ownership of a data domain.

1. Definition

   Data ownership is achievable by establishing clear definitions. Top data management is responsible for creating the correct definitions of data, analytics, and other pertinent parameters. Essentially, all members of the organisation should be on the same page and talk in the same language. This is the best way to understand and implement data management protocols.

2. Authorisation

   A good data literacy culture inculcates proper authorisation practices. For instance, who will be authorised to deal with private and sensitive data during e-discovery activities? The answer should fall under DataOps methodology. This method includes a hierarchy of roles and mindset being established that regardless of the location of the data, whether it is from another DataOps framework, data integration platform, or data catalogue, the authorised person knows the origin and destination of the data.

3. Management

   In management, there is regulation involved, such as ethical AI and privacy considerations, to protect the data integrity of your organisation. Data management is also the measurement of powerful tools in data analytics. Establish rules on how to measure your system if it is indeed delivering the analytics as promised, and the data analytics are operating properly.

4. Stewardship

Stewardship is taking responsibility for data and analytics. For instance, an organisation that operates in highly regulated environments where data quality and protection are paramount. There is always the possibility that data quality doesn’t meet the standards. Hence, somebody should be positioned to clean it up.

Next Steps:

• All large, and mid-sized organisations with a need for rapid innovation, should consider the CDO role. Like the Chief Information Security Officer (CISO) role, the CDO may be merged with the Chief Information Officer (CIO) in smaller to mid-sized organisations. No matter who takes on the CDO role, they should adopt an offensive data strategy, moving the organisation from viewing information from a historical perspective, to leverage data for prediction and decision-making.
• CDOs (or whoever is taking on the CDO role) should prioritise the four agendas outlined in this paper for 2021 through to 2025.