Data governance enhances business engagement, shared understanding, focus, and alignment, bringing an ever-increasingly disconnected data environment together, and providing data value optimization across many EDM initiatives.
Often, when people think of data governance, they think of data quality management. It’s true – data governance enables data quality management. However, there are many enterprise data management capabilities that can be so much more successful with a robust data governance operating model in place. Below, I identify several EDM initiatives and describe how data governance supports them.
- Metadata management
- Regulatory compliance
- Master data management
- Data quality management
- Data lifecycle management
- Overall enterprise data management
First of all, what’s a “robust data governance operating model”? The foundational elements for a data governance operating model include confirmed executive sponsorship, a defined organization with roles and responsibilities (and accountabilities) assigned, a charter that defines the scope of the data governance capability (which could expand over time), communication and training plans defined and developed by role, and a set of consistent policies and procedures that describe the people, process, and technology aspects of the data governance program in scope.
The data governance operating model can support the capture and management of both business and technical metadata. I’ve talked to people across so many companies who have indicated that they don’t know where their data resides and what it means. Technical metadata can be created using many different tools; ideally, one that can provide a robust data catalog and lineage. This information is useful, but the business definition of that data is a critical need.
The data governance office can coordinate the development of common business language and even a business glossary across impacted business areas, leveraging the data domain owners and business data stewards, and ensure that this information is integrated with the technical metadata for a full view of enterprise data, no matter where it resides. The value that this effort can bring to an organization is huge.
Data privacy regulations such as GDPR and CCPA have exacerbated the need for enterprise-wide regulatory compliance. A data governance operating model enables many aspects of regulatory compliance. For example, GDPR has highlighted the need for data privacy compliance. Data subjects can ask for the details of any information you have on them. They can ask that you pass along their data to another organization, or even that you permanently delete their personal information, particularly when you no longer have a defined need for it.
A data governance operating model will naturally identify the data environment and landscape through metadata. Developing a data framework through the data governance program can then enable business process mapping, risk analysis, and classification of data quite readily. Policies and procedures that support compliance can be developed by the data domain owners via the council, and the data governance office can develop appropriate training and ensure appropriate documentation resides in the data governance knowledge base.
Master data management/reference data management
Many organizations begin an MDM/RDM program with no data governance operating model in place. This situation creates many issues, since all aspects of mastering data require business input, especially ongoing management. While the MDM program includes developing and managing the technical capabilities, enabling entry of business rules to master the data and creating notifications and workflows for entry of and changes to business rules, data governance identifies the roles and responsibilities, decision rights, business requirements, rules & definitions, and standardized policies and processes to support MDM.
The business data stewards play a key role in master data management by providing business requirements and managing golden record attributes, hierarchies, data classifications, etc. to ensure that rules and corrections are applied within MDM or source applications to continue good quality.
Data quality management
Data governance and data quality naturally complement each other, whether it be associated with new development or ongoing monitoring/maintenance. A robust data quality management program entails data profiling, defining business rules to ensure “correct” data, implementing those business rules, monitoring critical data items on an ongoing basis, and ensuring corrective action takes place to continue good quality data.
A data governance office is a great area to coordinate and ensure the continuation of a data quality management process. Many of the activities described above are ideally performed by the business data stewards and technical data stewards. Without a data governance program, it’s difficult to find anyone to “own” the ongoing process to ensure continued high-quality data.
Data lifecycle management
The data lifecycle is defined in many ways. To put it simply, data is (1) created, (2) stored, (3) used, (4) archived or destroyed. Data governance facilitates the development of policies and procedures to support this lifecycle.
For example, Data domain owners, in collaboration with appropriate areas of the company (risk, legal, compliance, etc.) can define policies on data storage, data architecture, data standards, data quality, data classification, data access, data use, data sharing and data retention (to name a few!). The data governance office can then work with the data domain owners to identify the appropriate monitoring metrics once these policies have been implemented. Without a data governance operating model, coordinating these requirements and ensuring compliance is a complicated endeavor.
Overall enterprise data management
Self-service data management is becoming the norm in many organizations. This means that, not only are business users creating their own analytics and reports, but they may also be manipulating data and adding in their own business rules in individual data siloes.
This process of data discovery that most organizations desire frequently results in the creation of new data sets. If an analyst is adding inputs into the new data set and sharing the resulting content as well, then how can an organization ensure alignment with the original sources of the data?
Data governance can help with that! One of the goals of many organizations today is to empower the user and support the production of “new data.” Once created, an effectively designed Data Governance operating model can ensure that the metadata, e.g. owner, business rules, data used, calculation methodologies, how used, workflow, etc. is documented and that the new data sets are added to the organizational metadata inventory. Once in the inventory, it can be discovered and used by others in the organization with confidence.
To summarize, data governance enhances business engagement, shared understanding, focus and alignment, bringing an ever-increasingly disconnected data environment together, and providing data value optimization across many EDM initiatives. So, start with data governance. It’s the start of something wonderful!