Data governance framework: How to generate value from your data

Businesses generate a huge amount of data as they operate. Gaining insights, building strategy, and generating value from that data requires analytics skills and tools. And to maximize the value of data, businesses must plan for how to use and manage their data by implementing a data governance framework—an overarching model of how they will collect, manage, and archive their data.

What is a data governance framework?

A data governance framework is a collaborative model for managing enterprise data. The framework or system can set soft guidelines or firm boundaries around data creation and manipulation. Often a data governance team is established to ensure proper data use, data quality, and policy compliance.

Executing a data governance framework impacts your entire data management process, including architecture analytics and data models. Proper execution makes it easier to make smarter decisions, faster.

A successful data governance framework is one that fits the organization’s resources. It should adopt data governance best practices such as data quality, data security, metadata management, and monitoring and management.

What makes up a data governance framework?

The Tableau Governance Models describes three types of models: centralized, delegated, and self-governing data governance. Each defines options for varying levels of security, compliance, and management defined in part by the number and roles of the people involved in the collaboration.

Centralized data governance

In a centralized model, one authority, typically an IT team, owns and authors data sources and dashboards for various business groups in a one-to-many manner. A small set of “creators” passes data to many “viewers.” This is necessary when controlling highly sensitive data. For example, Providence St. Joseph Health manages thousands of patient medical records using solutions like Tableau to govern and analyze its sensitive data.

Delegated data governance

In a delegated governance model, additional roles are brought in from outside of the central authority. Site administrators or data owners are chosen and may have direct access to data sources. Content authors also have access to certified published data sources to ask and answer business questions. Access is delegated down for more collaboration.

Self-governing data governance

The most decentralized and distributed model is the self-governing model, which is more collaborative between IT and the various business groups utilizing data. Ad hoc reporting and additional access to data allow more teams to own and execute business reporting and decisions.

Why do organizations need a data governance framework?

Any organization collecting data and executing data analysis needs to be able to decide how to manage data. Showing value from the data, minimizing cost, managing risk, and complying with ever-growing legal requirements plays into these decisions. Organizations need to reach a consensus on how to make strategic decisions based on what they know and what has proven successful. They need to do more than manage data; they need a governance system that sets guidelines for engagement.

Small organizations or ones with simple data environments may be able to meet goals through a more informal governance model. Larger organizations or ones with more complex data or compliance environments typically need to organize and align with a more formal data governance framework.

What are the steps and considerations to successfully implement a data governance framework?

Like most business processes, data governance can’t be put in place then left to run unattended. The model must evolve in response to new information, challenges, and insights. A successful data governance framework has to become part of the organization’s culture, and the first step is gaining buy-in from necessary stakeholders and leaders.

Once your organization is aligned, the Tableau Blueprint Planner is a step-by-step guide to becoming a data-driven organization. It assists you with deploying and operating a data governance framework and will guide you through key decision points during the deployment process.

Steps to consider when implementing a data governance framework

Each step is critical to the success of the process. Ensuring all pieces are correctly placed starts with identifying the information you want to use to build your framework.

Define your analytics strategy

Start with your organization's strategic initiatives and KPIs, metrics, or outcomes, because you cannot manage goals if you don’t have the measurement framework in place. Knowing where you want to go and how you measure your success gives you the guide rails needed to meet your goals.

Identify members of the cross-functional team

Document who is responsible for understanding current/future state capabilities, challenges, and goals; set a vision with a project scope, prioritization, and success measures; hold regular meetings; gather/respond to feedback; and document value. Team roles might include:

  1. Executive sponsor, who sets the vision for modern analytics, aligns projects to transformational initiatives, nominates staff for advocacy roles, and ensures accountability.
  2. IT sponsor, who is responsible for data governance installation, configuration, and maintenance; partners with business leaders and SMEs; enables secure governed data access; and transitions content authoring to the business.
  3. Analytics sponsor, who implements the vision for modern analytics, ensures the availability of data and content, establishes education plans and learning paths by organizational job functions, facilitates communication throughout the user community, and aggregates business value achieved.
  4. Line-of-business sponsors, who advocate for data-driven decision-making within their respective teams, promote content authoring and governed access, support content, encourage collaboration and sharing, and document business value.

Document your current enterprise architecture

To execute a strong and collaborative data governance framework, it’s important to understand how systems interoperate and then work to incorporate them into your new model.

Define and answer questions for the following:

  • Hardware platforms: Where will you deploy? On-premises, public cloud, or hosted?
  • High availability/disaster relief: Is your server mission critical, requiring high availability?
  • Security: What is your enterprise authentication protocol?
  • Network: What is your policy on gateway ports over HTTP/HTTPS? Are there any port restrictions?
  • Operations: What are your enterprise software management tools? What processes exist for backup/restore?
  • Client software: How is client software deployed (desktop, prep, mobile)?
  • External services: Will you be integrating external services (R Server, Python, MATLAB, WMS)?
  • Licensing: What is the license type (role-based subscription, core, embedded)?
  • Data: What file-based sources of data will be used (network folder access)?

Understanding your enterprise architecture will help you identify sources of data that are important to job functions, prioritize which of these data sources will be certified and available at launch, and plan end-user training needs.

The Tableau Blueprint will guide you as you build a data governance framework for your organization. Whether you’re a small company or a large corporation, you can create and utilize a data governance framework customized to your organization’s needs for agility and proficiency by utilizing data to drive strong and strategic business decisions based on quality data.