any enterprises often face challenges regarding how data should be regulated or processed. Throughout the data transformation process, a vast amount of knowledge—spanning various roles and technical requirements—is accumulated, making it difficult for other stakeholders to quickly understand how that data should be utilized. To ensure data is documented and passed down in detail, many companies are expanding their “Data Catalogs” into “Knowledge Catalogs.” In this transition, the contributions of “Multiple Roles” within the data governance process become particularly vital.

Generally, when a factory manufactures a product, it provides an instruction manual for consumers who may lack prior experience with the item. The consumer does not need to participate in the manufacturing process or think about how to optimize the product; they simply follow the manual to use it. Data, however, is closer in nature to “raw material” than a finished product. It is difficult to create a single, exhaustive manual for its use. When data passes through different processing procedures or technologies—combined with various design specifications and interface presentations—it can be transformed into diverse data products. Consequently, how data is recorded and applied throughout this journey becomes the greatest pain point for users. This is where the involvement of multiple roles in data governance becomes critical. During the data production process, these diverse roles provide detailed annotations and share application-specific knowledge, expanding the data catalog into a knowledge catalog through documentation and heritage. These individuals are not just end-users; they are active participants in the ideation and creation of data products.

Data governance carries the collective knowledge required by numerous positions, roles, and technical functions within an organization. If this knowledge or existing case studies of data products are missing, it becomes nearly impossible for other users to brainstorm or understand how to leverage that data. Therefore, successful data governance requires continuous iteration and consistent execution. It relies on the collective participation and discussion of multiple roles, with the data governance steering body then publishing the key priorities. Furthermore, as corporate data grows at an exponential rate, it is challenging for teams to apply the same level of optimization, definition, or knowledge supplementation to all data. Within the constraints of time and cost, enterprises must establish stricter workflows and communication details to ensure that data records remain useful for future purposes.

The essence of data governance is “governance” itself. There are many ways to achieve governance goals, each with its own set of standards, pros, and cons. Consequently, enterprises must be clear about their expectations for data-driven decision-making to advance relevant governance measures.

Through the participation and contribution of multiple roles, an organization can: