How to implement Computational Data Governance to achieve measurable data maturity progress
In today’s data-driven world, effective data governance is essential. Organizations must ensure data quality, security, compliance, and accessibility to drive informed decision-making, maintain customer trust, and meet regulatory demands. As data volumes grow and AI-driven analytics become the norm, governance provides the guardrails needed to prevent errors, biases, and breaches. With rising cyber threats and evolving privacy laws, strong data governance turns data into a strategic advantage rather than a liability.
We’ll first explore the history and current state of data governance, as observed in various organizations over the years. In a subsequent article, I’ll outline how to effectively implement computational data governance. We’ll illustrate how to measure effectiveness and penetration by defining maturity levels and end by learning how to manage these levels to continuously achieve measurable data maturity progress.
Data Governance, a necessary evil?
Despite its critical role, Data Governance often carries a negative perception within organizations. From a business perspective, governance is frequently seen as a bottleneck — something that slows down innovation and decision-making. Business leaders may view it as restrictive bureaucracy that adds unnecessary complexity rather than enabling agility and data-driven growth.
From a management perspective, many executives perceive data governance as a costly compliance exercise rather than a value driver. Without clear ROI, it can seem like an overhead function that drains resources without directly contributing to revenue or competitive advantage.
From an engineering or IT perspective, developers and data engineers often see governance as an obstacle to speed and flexibility. Rigid policies, extensive documentation, and access restrictions can make data harder to use, slowing down development cycles and hindering rapid experimentation with AI, analytics, and automation.
These perceptions stem from traditional, rigid governance approaches. However, modern governance strategies — such as Data Governance by Design and Computational Data Governance — aim to embed governance seamlessly into workflows, shifting it from a constraint to an enabler of efficiency, security, and innovation.
Data Governance by Design: building a proactive foundation
Data Governance by Design embeds governance principles such as quality, security, privacy, and compliance directly into data systems and workflows from the start. Instead of enforcing rules after data is collected, this approach ensures governance is an inherent part of data architecture, reducing risks and improving efficiency in an increasingly regulated and data-driven world.
It is, however, mostly descriptive, providing documentation and filling up catalogs. Although valuable, the quality of the information depends on the accuracy and willingness of the human providing it, on keeping it up to date, and on providing meaningful use of it. Often, one should be “ticking the boxes” as part of a process, do so once, and there you are, you have achieved the holy grail of Data Governance by Design.
Why Business, not IT, should own data
Shifting data ownership from IT to business is a crucial step organizations often take. It is supposed to be a cornerstone to improving data maturity and data governance. Business teams understand data’s value, context, and quality needs, ensuring it is accurate, relevant, and aligned with strategic goals. This ownership drives better governance by embedding accountability where it matters, reducing reliance on IT for data decisions.
As a result, organizations move from reactive, IT-driven governance to a more agile, business-led approach that enhances data quality, compliance, and innovation. But then again, are we not just ticking the boxes? How is a business department supposed to execute on this ownership? How is this supported by IT? What means and tools do they have at their disposal?
Do we have real, actionable data ownership by business? Or does it only exist on paper?
The role of Computational Data Governance
While Data Governance by Design sets the foundation, and business data ownership is often established, Computational Data Governance takes it further by automating governance policies using process automation and rule-based systems, or even AI and machine learning for the not so faint of heart. It ensures continuous enforcement and adaptation at scale, minimizing human intervention by automating processes and supporting workflows. Together, these approaches shift governance from a reactive burden to an integrated, intelligent system that allows measuring performance and improvements, and evolves with increasing data maturity needs.
Federated, computational data governance is one of the guiding principles of the Data Mesh. By now, most organizations realize that a Data Mesh, and as such the federational aspect as one of the key promises, is for most organizations often more than one bridge too far.
The battle between a centralized and a more decentralized approach has by now slowly faded away. But the computational, automated approach to data governance should be the next step, the goal “an sich.”
Why, you ask? Because it’s not only about adopting effective data governance in an organization, but also using it to improve the overall data and data management maturity.
But first, how do you approach Computational Data Governance?