Data governance is the overall management of the availability, usability, integrity, and security of data used within an enterprise. A sound data governance program is driven by a governing body or council, which encompasses stakeholders from across the business, including data stewards. It includes a defined set of procedures, policies, and a plan to execute on those procedures.
Realization of corporate data governance is largely motivated by a desire to improve business operations and performance by gaining better oversight and management of corporate information. While a data governance program institutes policies and processes designed to produce more accurate and consistent data throughout an organization, it primarily becomes the job of the data steward to put those policies and processes into practice by ensuring compliance. It’s through governance and enforcement of said policies where you are ultimately supporting business process integrity, which in turn drives positive business outcomes.
Data Quality Policies
Data policies allow users like data stewards to define thresholds, and monitor breaches and deviations in the quality of the existing data as well as incoming data.
Data quality policies apply Metrics on Datasets to measure the quality of data. Thresholds define when users must be notified.
Policies enable data stewards to proactively monitor data. Data stewards can define policies to ensure data completeness, uniqueness, accuracy, and more.
For more information, refer to the Data Policies topic in the Data Governance documentation.
Existing Data versus Incoming Data
Existing Data Policies evaluate data that exists in STEP each night. Incoming Data Policies only evaluates the incoming data from an inbound integration endpoint of the Merge Golden Record type, allowing early warnings if the source system starts sending bad data.
Data Quality Dimensions
With Data Quality there are dimensions that need to be considered which are key cogs in driving the definition of Data Quality Policies.
Note: Data Quality dimensions are not to be confused with language dimensions, country dimensions, etc., that are platform-specific concepts.
Data Quality policies help organizations to ensure that data quality complies with the business' expectations.
Data quality policies use logical metrics on entity data to test the quality threshold. These thresholds show a simplified view of the metric performance. If data quality does not comply with the policy, a data policy breach is recorded. These policies update when the data quality returns to normal expectations.
With these policies, data stewards are able to proactively monitor, control, and maintain entity data from within MDM. Data stewards can build policies to ensure data completeness, uniqueness, accuracy, and more.
Examples of data quality dimensions are:
- Accuracy - Is the data verified, accurate and up to date? Our 3rd party integrations can now verify aspects of party data using the very latest trusted reference sources.
- Completeness - What data is missing or unusable? Is all the necessary data present?
- Timeliness - The degree to which data represents reality from a required point in time. Is data available at the time needed?
Metrics define the specifics of how to measure data quality.
For use cases with data governance concepts, refer to the ACME Holding Group Example Case topic.