The fourth industrial revolution (FIR) has invariably given rise to a data economy. Any business necessarily invested in it faces the imperative to institute and maintain data governance principles and policies as an integral aspect of their participation in the data economy. If anything, the FIR has made it apparent that data is a keystone for nearly all businesses, regardless of size. Furthermore, the exponential increase in data production has given rise to the growing prominence of data analytics, which is the main tool for extracting insights from data. In fact, there have been various core industry changes to fundamental business practices ‘instigated’ by data and data analytics, as demonstrated in a study by McKinsey Analytics (2018). The following chart is one such demonstration.
Source: (McKinsey Analytics 2018, p43)
If anything, these illustrations underline the importance of data management. Data management is succinctly defined as the practice of undertaking disciplines relating to managing data as a valuable resource. Such disciplines particularly relate to key attributes of data including (not exhaustively) quality, availability, usability, security, consistence, and integrity. Given the growing prominence of data as a valuable resource in business, it is only imperative, at the least, that data be managed with the highest of priorities if desired business goals are to be achieved with the help of data. In fact, unmanaged data can have the adverse effect of undermining progress towards the achievement of business goals.
Data management invariably comprises a plethora of concepts that are operationalised to make up the necessary disciplines to be undertaken as part of the data management process. This is owing to the multitude of dimensions of the data being produced by the different kinds of businesses and industries. As such, it is necessary for each business to ensure that it is well versed with the data management disciplines necessary for its type of data and industry. However, given the continuing nature of the data management process, it is important to establish a set of data management principles that will enable each business to focalise the fundamental aspects of data as a valuable resource in attaining its business goals (Smith 2016). The following are some of these principles, as adapted from the principles suggested by Smith (2016) and Pal (2019):
- Recognising data as an asset: It is important to appreciate data as an asset with measurable value. This might seem trivial, but a company culture with an appreciation of data as an asset embedded in it can have an immense positive effect on business operations. The value of data culminates in the development of insights used to complement decision making and developing business strategies. The importance of this cannot be overemphasized. Considering this fact, it is important to ensure that data assets are defined, controlled, accessed, and used in a process-driven manner to maintain the integrity of data as an asset. This way, management can have confidence in using data-driven insights to make important decisions.
- Implementing data standards: The data management process in any country is most likely governed by rules and regulations pertaining to data stewardship most likely set by the country’s central statistical office. In South Africa, Statistics South Africa (2010) developed the Statistical Quality Assessment Framework for this purpose. If the business is operational on a global scale, it should also adhere to other global data management standards such as those set by the International Monetary Fund (2007). Security is one such standard, with businesses required to ensure privacy, confidentiality and appropriate access by defining clear rules and guidelines for all individuals handling data to adhere to. Quality is another standard. Data quality has to be consistently managed throughout the data’s entire life cycle, with quality checks instituted to ensure acceptable levels of quality as defined by set standards.
- Developing and maintaining consistent metadata definitions: Metadata is important in enabling a good understanding of all aspects of the data when the data is handled within the business, or by an external stakeholder. This is also important in establishing a culture of understanding and appreciating the value of data.
- Implementing support structures for integrating data from multiple sources: This is an important quality standard which enables the extraction of in‑depth insights from data. Integrating data from multiple sources, including historical sources, can help discern trends and processes that would otherwise not be apparent.
The list of these principles goes on and it is the responsibility of all data custodians to familiarise themselves with those that are most relevant to their business. Organisations typically express these principles in mission statements and organisational aims and objectives. Over and above all these principles, it should be appreciated that data management is a continuous process which should be continuously improved. Each business should identify the core principles it needs to emphasise in relation to the data it produces and uses. These should then be used to develop policies and guidelines to be strictly adhered to in order to maintain good data management practice.