AI glossary Strategy & Business
What is data governance?
← All glossary termsThe ownership, quality rules, lineage and policies that decide who can use which data, and how, so it is fit for analytics and AI.
Explained
Data governance is the set of roles, decision rights, policies and controls that determine who owns which data, who may use it and for what, how it is classified, retained and protected, and how its quality is assured. It connects privacy, security, master data and, increasingly, model risk. Good governance is a prerequisite for compliant AI, reliable features and explainable results, because a model can only be as trustworthy as the data it learns from or retrieves. It is an ongoing function, not a one-off project.
What it covers
- Ownership and stewardship: a named owner for each important dataset, and people responsible for its day-to-day quality.
- Classification: which data is personal, confidential or regulated, and what rules apply to each class.
- Access and use: who can see and use which data, for which purposes, with what approvals.
- Quality: definitions, checks and remediation for accuracy, completeness and timeliness.
- Lineage and metadata: where data comes from, how it is transformed and where it is used.
- Retention and deletion: how long data is kept and how it is removed.
Why AI depends on it
AI systems learn from data, retrieve from it and act on it. Errors, gaps and biases in that data become errors, gaps and biases in the results, often in ways that are harder to see than in a report.
Governance answers the questions that come up in every AI project: can we use this data for this purpose, is it accurate and representative, who has to approve, and can we show later which data a model used. Without those answers, projects either stall or move ahead with risks nobody has assessed.
Data governance and the EU AI Act
For high-risk AI systems, the EU AI Act sets specific requirements on data and data governance. Training, validation and test datasets must be subject to governance and management practices suited to the system's purpose, covering among other things the origin of the data, preparation steps such as labelling and cleaning, and the examination of possible biases.
This builds on existing obligations such as the GDPR when personal data is involved, but it applies to data quality as such, whether or not the data is personal.
Frequently asked questions
What is the difference between data governance and data management?
Data management is the overall practice of collecting, storing, integrating and using data. Data governance is the part that sets accountability, policies and decision rights over it. DAMA's body of knowledge places governance at the centre of data management, guiding the other areas.
What are the main pillars of data governance?
Most frameworks include ownership and stewardship, data quality, metadata and lineage, security and access, privacy and compliance, and lifecycle management. The exact split varies, but all of them assign responsibility and set rules that are checked.
Who is responsible for data governance?
Responsibility is shared. Business owners decide on their data, data stewards look after its quality, IT and data teams implement the controls, and a governance body or chief data officer sets policy and resolves conflicts. Governance fails when it is left only to IT.
How does data governance relate to AI governance?
AI governance covers the whole lifecycle of AI systems: purpose, risk, models, oversight and accountability. Data governance is one of its foundations, covering the data those systems use. The two need to be connected, for example so that a model's training data can be traced and approved.
How do you start a data governance programme?
Start with the data that matters most for a few priority use cases, name owners for it, agree definitions and quality rules, and put basic access controls and a catalogue in place. Expanding from real needs works better than designing a complete framework before anything is used.
Decision guides that use this term
How to put it to work
Where it shows up at Thinkia
Further reading
Related terms
Sources
- DAMA International · Data Management Body of Knowledge (DAMA-DMBOK)
- EUR-Lex · Regulation (EU) 2024/1689 (Artificial Intelligence Act), consolidated text
- AI Act Explorer · Article 10: Data and data governance
- ISO · ISO/IEC 38505-1:2026 Governance of data, Part 1
Updated: 2026-10-21