Data governance programs are as unique as the companies that implement them. However, the frameworks for data governance programs are actually pretty similar for each implementation. There are certain foundational components on which governance is built. I'm going to briefly describe each component and then describe how that component looks when it's being properly managed. In other words, we're going to describe what "right" looks like.
The six components of a data governance framework are:
- Organization - administers activities and provides a responsible network of resources to deliver governance capabilities.
- Policies, principles and standards - refers to information management guidelines and principles for enforcing data standards and governance procedures.
- Processes and practices - establishes guiding principles for how policies and processes are created, modified and implemented.
- Metrics - establishes measures for monitoring information/governance performance and actions to continually improve enterprise data quality.
- Data architecture - includes enterprise data standards, a business information model, metadata dictionary, as well as security and privacy measures.
- Tools and technology - supports common information exchange solutions, workflow and business rules, as well as user presentation.
There is a right way to implement each of these components. It's not really up for debate - either it works, or it needs improvement. It's pretty quantifiable.
The first component is organization. When data govrnance is implemented effectively, there is a chief data officer in charge of the effort to provide strategic leadership, executive direction and oversight to drive sustained alignment with business priorities and compliance with regulatory mandates. There are also clearly defined roles and responsibilities, including those for data ownership and stewardship. There are data standards in place that set overarching governance policies and serve as a forum for resolving conflicts.
For the component dealing with policies, principles and standards, we need to see that common data standards have been established and implemented across lines of business for key data subject areas such as customer, contracts, vendors, products, etc. The governance policies and procedures will look clearly defined, broadly communicated and understood throughout the enterprise. This helps reinforce data standards, particularly for the key data subject areas.
Data governance processes and practices will be right when key DG processes (such as those for change requests, compliance and exception reporting, issue resolution, etc.) are clearly defined, established and tested across the enterprise. Areas where business processes are impacted by data governance processes will have been identified and suitable actions will have been taken to address and mitigate those impacts.
When the metrics component is implemented effectively, the metrics will be measuring and providing information on what they're supposed to measure. Metrics should provide consistent, periodic measurement standards to effectively monitor data quality, compliance with data governance policies and standards, and the overall performance of the governance organization.
When the data architecture component is executed properly, there is a practical, usable data model in place, and all data quality and metadata management requirements have been identified and implemented. The implemented architecture should be extensible, scalable and robust, while being flexible enough to meet the requirements of an ever-changing IT environment. Finally, it should support key line-of-business functions at the enterprise level.
The final component, tools and technology, is effectively implemented when key business, information management and governance processes are automated across LOBs and business functions as much as is practicably possible. The tools and technology should also provide full support for governance support activities such as workflow and content management, as well as DG standards and policies.
The potential benefits that arrive when we do data governance right are enormous, especially in the long term. Efficiency will be improved via access to higher quality data for decision-making. LOBs are better able to coordinate their activities due to standardized processes and access to enterprise-wide data. This can provide substantial cost savings. Savings will also be achieved by reducing the number of IT applications and systems and standardizing the ones that remain.
Capacity will be improved due to better reporting and analytical capabilities arising from improvements to data quality and access. Regulatory compliance can likewise be improved due to standardized processes and data needed to maintain compliance performance. The culture of data ownership and stewardship that an effective data governance program will engender can also help maintain compliance. With benefits and opportunities like these, doesn't it make you want to really get it right?
Deloitte is not, by means of this article, rendering business, financial, investment, or other professional advice or services. This article is not a substitute for such professional advice or services, nor should it be used as a basis for any decision or action that may affect your business.
Jane Griffin is a Deloitte Consulting LLP partner. Griffin has designed and built business intelligence solutions and data warehouses for clients in numerous industries. You can follow Jane on Twitter at @janegriffin. She may be reached via e-mail at email@example.com.