data governance intelligent organization generative ai digital transformation

Mastering Data Governance: A Guide for Optimal Results

With digital transformation initiatives on the rise, organizations are investing more in Data Governance, a formalized practice that connects different components and increases data’s value. Some may already have established Data Governance programs for older Data Management systems (such as for controlling master data) but may lack control in newer technologies like training an AI to generate content and make need guidance in best practices to follow.

Steve Zagoudis, a leading authority on Data Governance, notes that a lack of awareness explains some of the disconnects in applying lessons learned from past Data Governance to newer programs. What’s more, Data Governance has a bad reputation as a drag on innovation and technological advancement because of perceived meaningless workflows. 

To turn around these trends, companies should embrace Data Governance best practices that can adapt to new situations. Furthermore, businesses must demonstrate how these activities are relevant to the organization. Using the tactics outlined below promises to achieve these goals. 

Lead by Doing 

With Data Governance, actions speak louder than words, especially regarding newer projects using newer technologies. Any communications from the top-down or bottom-up need to show how Data Governance activities align with business innovations. Try having:

  • Executives lead as engaged sponsors: “Executives need to support and sponsor Data Governance wherever data is,” advises Bob Seiner. Often, a data catalog (a centralized metadata inventory) can help guide executives on where to apply Data Governance. When implementing Data Governance, managers should communicate consistently and clearly about the approach, roles, and value of Data Governance. They need to emphasize that these aspects apply to new projects too. Moreover, senior leadership needs to visibly support and allocate resources – time, money, technology, etc. – toward data stewardship, formalizing accountability and responsibility for company data and its processes. 
  • Data stewards lead through information sharing: Data stewards typically have hands-on experience with company data. Consequently, these workers are a treasure trove of knowledge valuable to their co-workers, manager, and other organizations. Not only does this information exchange help others in the company learn, but sharing also activates data stewards and keeps them highly invested in Data Governance practices. With this advantage, stewards are more likely to extend their work to newer projects.
  • All employees lead by applying a company’s Data Governance best practices: All employees take care of the Data Quality and communicate when they have questions or helpful feedback. Business leaders should provide two-way channels for stewards to encourage Data Governance adoption among their departments and allow users to express their problems or ask questions.

Understand the “Why”

Business requirements change quickly as companies become more data-driven. For example, the metadata requirements previously used to describe application error data and set forth by Data Governance may need a different format to train a generative AI model to suggest fixes.

To keep Data Governance relevant, teams must create actionable use cases and connect the dots to the Data Governance’s activities. Out of this work should come a purpose statement defining success with the measurements and stories to show company project progress achieved from Data Governance.

Data Governance purpose statements help navigate the support needs of data products, ready-to-use, high-quality data from services developed by team members. To justify updates to Data Governance processes, business leaders should present new data products as a proof of concept and explain a roadmap to get to the changes. Consider integrating a few critical Data Governance activities and how they benefit the data product in the presentation.

By using the Data Governance purpose statement as a guide and building out solid use cases tied to data products, teams can understand the benefits of good Data Governance and the consequences of poor Data Governance. Furthermore, this messaging solidifies when it is repeated and becomes self-evident through data product usage and product maturity.

Cover Data Governance Capabilities

Before starting or expanding new projects, organizations must be clear about their capabilities to adapt to Data Governance activities. For example, if a software application needs to ship in three months, and three-quarters of the team must spend 90% of their time and money getting the technology running and fixing bugs, then Data Governance resources for metadata management through Data Governance will be scarce.

To get a complete picture, organizations usually assess where their Data Governance and its best practices stand today, addressing best practices and maturity.

Once companies have compiled feedback and metrics about their Data Governance practices, they can share recommendations with stakeholders and quickly check improvements and goals as they apply Data Governance. As resources fluctuate, business leaders might consider bringing Data Governance into project daily standups or scrum meetings to track and communicate progress.

As project managers and engineers help one another when blocked, they can note when a data product story with Data Governance activities has been completed. In addition, adding Data Governance to daily meetings can prompt team members to bring back Data Governance components that have worked in the past – data, roles, processes, communications, metrics, and tools – and reuse them to solve current issues. 

Implement a Well-Designed Data Governance Framework

A well-designed Data Governance framework provides components that structure an organization’s Data Governance program. Implementing such a framework means that Data Governance assures an organization of reliable data with a good balance between accessibility and security.

Over 60% of organizations have some Data Governance that is in the initial stages, according to the recent Trends in Data Management report. Existing Data Governance programs can take many different formats, including:

  • Command-and-Control: A top-down approach that sets the Data Governance rules and assigns employees to follow them
  • Formalized: Training programs constructed as part of an organization’s data literacy initiative to encourage Data Governance practices
  • Non-Invasive: A formalization of existing roles 
  • Adaptive: A set of Data Governance principles and definitions that can be applied flexibly and made part of business operations using a combination of styles

The best approach works with the company culture and aligns with their data strategies, combining choices and decisions that lead to high-level goals. 

Gather the metrics and feedback about Data Governance capabilities to understand what processes, guidelines, and roles exist and are working. Then, decide how many existing components can be used versus how much work needs to reframe the Data Governance approach. 

For example, a command-and-control construction may allow enough flexibility in a start-up environment with two or three people; however, as a company adds more employees, Data Governance may need to be reformulated to a non-invasive or more adaptive approach. 

Evaluate automation, such as a data catalog or Data Governance tools, regardless of the Data Governance framework chosen. Ideally, companies want automation that empowers workers in decision-making and adapts as necessary to the Data Governance purpose.

Develop an Iterative Process

To adapt, companies must develop an iterative process with their Data Governance components. This tactic means flexibility in adjusting goals to get to the Data Governance purpose.

For example, a Data Governance program’s purpose ensures Data Quality – data that is fit for consumption. Initially, Data Governance members discuss critical data elements around a data model built by a team. 

Should this task lead to unresolved disagreements after a sprint, business leaders can try shifting gears. Shelve the debate and focus on connecting terminology to shared automation tools the members use.

Specific Data Governance processes may need updates as data moves between older and newer technologies. These cases may need new Data Governance stories for sprint planning and execution. Once an organization finds out what works over a few sprints, the team can repeat these activities and consistently communicate why and how the workflow helps.


Because business environments change rapidly, Data Governance best practices must be adaptable. Gartner has estimated that 80% of organizations will fail to scale digital business because they persist in outdated governance processes. 

Versatile Data Governance activities require engagement from all levels of the organization and especially sponsorship from executives. Flexibility comes from understanding the purpose behind Data Governance activities and knowing Data Governance capabilities, to be able to use what works to the best extent.

Data Governance needs implementation through a good framework that includes automation. In addition, any software tools supporting Data Governance need evaluation on how well they match the Data Governance’s purpose. 

Data Governance best practices must work in iterations to become agile in changing business contexts. Businesses should plan on modifying the Data Governance controls used today as new technologies emerge and business environments evolve.

Author: Michelle Knight

Source: Dataversity