Doing business now means generating, storing, and accessing more data than ever before. An effective data governance model is essential for keeping your organization effective and on the right side of regulations.
Current projections indicate the world will create and consume more than 181 zettabytes of data — that’s about 181 billion terabytes — by 2025, compared to just 2 zettabytes in 2010. Businesses in the current era of Big Data must be proactive about who in their organization is responsible for their share.
To be effective, data governance models and data governance operating models cannot be the sole purview of an organization’s IT department. They must be observed by all data stakeholders at each level, requiring interdepartmental commitment and communication. In return for the effort, every part of your organization will benefit from accessible and reliable data.
Jump to a section…
What is a data governance model?
How to design a data governance model for your business
Data governance model examples
Top-down data governance
Bottom-up data governance
Collaborative data governance model
Stay on track with a data governance maturity model
Learn more about how to keep you and your business’ information secure with the Symmetry Systems’ guide, What Is Data Governance?
What is a data governance model?
A data governance model is a set of rules, roles, and processes that organizations apply to their data management practices to ensure their data is accurate, secure, and compliant. Every part of a business that contributes and uses data should be represented in the data governance process. This stands in contrast to data management, which is the specific set of IT processes used to create, store, transmit, archive, and destroy data across its life cycle.
Internally, a data governance model must account for the specific needs of the organization, which may vary greatly depending on your industry and the size of your company. Externally, data governance is essential for compliance with business-wide regulations, such as the EU’s GDPR, as well as industry-specific requirements, such as HIPAA for companies involved with healthcare in the US.
While specific data governance models vary greatly from company to company, here are three universal concepts:
- A data governance model dictates organization-wide policy, requiring buy-in from all data stakeholders.
- A data governance operating model dictates specific steps employees on the ground should follow when creating and handling data.
- A data governance maturity model tracks an organization’s progress in implementing an effective DGM, and what kind of results it should expect at each stage.
Multiple approaches to data governance models may be feasible, depending on your organization’s needs.
How to design a data governance model for your business
The cross-disciplinary nature of data governance means it will take both time and good communication to establish. It is as much a process of establishing contact points and lines of communication as it is about implementing specific technical processes.
However, effective data governance can have a concrete impact on businesses; a 2019 survey found that 30 percent of employees’ time on average was taken up by non-value-adding tasks required by poor data quality and availability. Better data governance means employees spend less time searching for good data, and more time using it.
The first step in setting up a data governance model is establishing data owners: people who can make decisions about data and are responsible for its accuracy and relevance, and who can ensure employees working underneath them are similarly compliant. Larger organizations should seek out the expertise of ground-level data stewards and subject matter experts who use data on a day-to-day basis. Finally, a data governance committee composed of stakeholders across the organization will ensure the processes meet everyone’s needs and help guide integration efforts.
From there, you can determine the scope of the data governance model you want to start with — if you’re starting from scratch, you shouldn’t try to encompass the entire company’s data all at once. Applying your data governance model to government-regulated data is an excellent place to start, since it already has external criteria which must be met.
Data governance model examples
Broadly speaking, data governance models fall within three approaches.
Top-down data governance
In this approach, a small team of data professionals control the creation and use of data through set methodologies. It ensures that data is clean, reliable, and up to date by limiting access to individuals whose primary responsibility is data management. While effective for smaller organizations, this approach can create bottlenecks for larger companies. Furthermore, employees may feel compelled to circumvent the proper data channels so they can get their work done, invalidating the purpose of data governance.
Bottom-up data governance
This method imposes few restrictions on the ingestion of raw data, opting instead to apply structures and filters after its creation to ensure quality and consistency. While this method eliminates the bottlenecks of running all data through specific team members, it also makes it much more difficult to ensure quality. The efficiency saved in creating and accessing data may be lost later by the time required to make sure the data you have is what you actually need.
Collaborative data governance model
This is the middle road between top-down and bottom-up data governance models. Spreading responsibility for managing data sources across the organization, while retaining well-defined roles for data owners and stewards, allows a company to approach each task with both flexibility and consistency. Maintaining these distributed responsibilities across departments and projects requires ongoing effort and communication, but is likely to yield the best results for most applications.
For more data governance model examples to consider, you may wish to consult Microsoft’s guide to data governance.
Stay on track with a data governance maturity model
Effective data governance requires a broad effort across an organization, and yields similarly broad results. This makes it difficult to use standard KPIs to track the progress of its implementation. Instead, you should seek out a data governance maturity model to track accountability and results. The state of Oklahoma’s Office of Management & Enterprise Services’ data governance maturity model, which is based on industry-standard tools from Stanford and IBM, is an excellent place to start.
As organizations implement data governance to better protect and understand their data, the need for strong data security at every level becomes ever more clear. Symmetry Systems helps security teams protect sensitive data in the cloud, providing visibility into where your sensitive data is, who has access to it and how it’s being used to both proactively identify potential issues before they can be exploited and your blast radius in the event of a breach. If you’d like to learn more about how Symmetry DataGuard can fit into your data governance methods, please contact us today to learn more.
Share this entry