I’m often asked if there is a difference between data governance and Data management. The answer is yes—but they are related.
data governance is the definition of organizational structures, data owners, policies, rules, process, business terms, and metrics for the end-to-end lifecycle of data (collection, storage, use, protection, archiving, and deletion). Data management is the technical implementation of data Governance.
Data governance without implementation is just documentation. Enterprise data management enables the execution and enforcement of policies and processes.
Here’s how I described the difference to my father, who worked in the construction industry for more than 50 years: data governance is the blueprint for a building, and data management is the physical construction of the building. Without data management, there is no physical building. And while you can construct a building without a blueprint (data governance), it will be a less efficient and less effective activity, with a greater likelihood of problems down the line.
What is Data Governance?
Let’s take a closer look at some aspects of data governance.
People: People are critical to data governance because they are the ones who create and handle the data, and ultimately benefit from well-governed data. Examples include the subject matter experts in the business who can determine both standardized business terms for the Organization, along with the levels and types of quality thresholds required for different business processes. Data stewards are responsible for remediating data quality issues. IT people are responsible for the architecture and management of databases, applications, and business processes. Legal and security people are responsible for data privacy and protection. And cross-functional leaders, who comprise the governance board or council responsible for resolving disputes between different functions within an Organization.
Policies and rules: If policies define what, rules define how. Organizations use a wide range of policies and rules across processes and procedures; common categories include consent, quality, retention, and security. For example, you might have a policy that states consent for processing must be obtained before personal information can be used. One rule might define the consent options (like billing, marketing, and third-party sharing) that people must select when personal data is being collected. And another rule might define that marketing consent must be confirmed before sending a promotional offer to a customer.
Metrics: What gets measured gets managed. Common technical metrics include things like the number of duplicate records in an application, the accuracy and completeness of data, and how many personal data elements are encrypted or masked. While these types of metrics help in the technical management of data, leading organizations are also looking to define how these technical metrics impact business outcome metrics.
For example, days sales outstanding (DSO) is a common business metric used by financial analysts and lenders to analyze the financial health of a company. If customer address data is incomplete or inaccurate, it will increase the billing cycle time and consequently increase DSO.
