Distribution of Responsibilities: We recommend the differentiation of the following three Data Governance levels:
Executive management level – comprised of a person with the top-level responsibility for organization’s data, including Data Governance, Data Quality, Data Architecture, Data Protection etc., usually being called “Chief Data Officer”.
Data Stewardship level – comprised of people responsible, from the process perspective, for the definitions, implementations, validations and quality of data sets, usually being called “Data Owners”.
Data Management level – comprised of people providing technology and IT services, including data architects, designers, developers, and various administrators, usually being called “Data Custodians”.
Integration with Business Processes: Since organization’s data is the “fuel” of its business processes and the quality of the data influences the business performance, we advise focusing on the Data Governance in relation to the business processes. Process Owners (a person responsible for a business process) represent Data Users, i.e. those people who formulate their expectations on data. We recommend the approach starting with the business processes, their Process Owners, leading towards critical data sets, and progressively establishing Data Governance from there.
Integration with Information Model: Despite the fact that “governance” of “anything” presumes that “anything” is somehow defined, we have seen several times that an organization does not have even a “data globe”, a high-level map, of its data when considering to establish some Data Governance. Thus, we advise to at first conduct a review of organization’s data and elaborate the “data globe”, describing, at a higher-level, or organization’s data with fundamental attributes and relationships, i.e. an Enterprise Information Model. On top of such a model, the Data Governance is built.
Integration with Data Management: Data Governance is the central topic of the famous DAMA Wheel. That is not just a matter of a visual illustration, it has a deeper message. Data Governance is a fundamentally prerequisite function for effective and efficient existence of all other Data Management functions. For example, no meaningful Data Quality Management can exist without Data Users (who say what they expect from data), Data Owners (who are responsible to reconcile Data Users’ expectations and manage that the data fulfill what is promised to the Data Users) or Data Stewards and Custodians, the executive workforce.
