It is great to see that data governance and data management are getting the attention they deserve in this CFA Institute Enterprising Investor article. To make it more tangible, I would draw parallels with investment governance and discuss data quality with the GIPS Compliance example.

Data governance is similar to any other governance, including investment governance, as outlined in the CFA curriculum. It starts with identifying the purpose of governance (e.g., ensuring high-quality data for sound investment decision-making and regulatory compliance). It then outlines the principles to follow (e.g., adopting an approach based on criticality of data), defines governance structures (e.g., board risk committee, executive risk committee, any other governance forums), allocates roles and responsibilities (e.g., data owners, data stewards), and outlines required processes (e.g. identifying critical data, tracing lineage, assessing controls and data quality, monitoring and reporting, managing data issues), standards and controls across the data lifecycle (e.g. data quality, privacy, security). I have seen the language of data risk resonate well in financial services. Whether it is due to loss aversion, recent regulatory action, or a broad understanding of the risk management frameworks that can be readily mapped to data risk, it helps the implementation of data governance and management.

High-quality data is required for anything an organization does, although the definition of "high" is use case- and stakeholder-specific. The most basic example is the Global Investment Performance Standard (GIPS) for Firms, which states that the inputs to performance presentations need to be "accurate", which in turn requires the implementation of data quality controls along the flow of data and across the lifecycle. Other data quality dimensions, such as timeliness and completeness, would also impact the accuracy of the performance presentations. It helps to think about the data risk and required controls through the lens of the data quality dimensions and the data lifecycle stage.