What is data governance?
Data governance is the combination of roles, rules, processes and standards with which a company ensures the quality, security, availability and compliant use of its data. It defines who is responsible for which data, how it is defined, and who may access what.
Also known as: data management rules · data stewardship
Where data governance is used
Data governance ensures that data is treated as a reliable corporate asset. It regulates how terms and metrics are defined consistently, who maintains and approves data, how data quality is measured, and how access and data protection are handled. In doing so, it builds trust in the data on which decisions are based.
In concrete terms, governance covers roles such as data owner and data steward, a shared understanding of terms (glossary and definitions), quality rules, access and permission concepts, and requirements for data protection and compliance, for example under the GDPR.
A practical example
Without governance, the same term often means different things in two departments, for example when revenue is meant once with and once without returns. Governance binds how such metrics are defined and makes reports comparable across departments.
In the dy Project AG data platform, governance was central: for a large construction project worth over 1 billion CHF with data from SQL Server, Excel and REST APIs, it had to be clear which data is reliable, how it is defined and who may see which views. The medallion architecture helped separate validated from unvalidated data.
Benefits & typical use cases
Data governance becomes important as soon as several teams access the same data or regulatory requirements exist.
- Consistent definitions so that metrics mean the same thing in every report
- Traceable data quality and clear responsibilities for maintenance and approval
- Regulated access and permissions, for example via row-level security
- Data protection and compliance, for instance under the GDPR
How it differs from related terms
The data strategy is the overarching layer; data governance is its operational rulebook. Data quality is a goal that governance secures. The technical platform such as a data warehouse or lakehouse implements governance, for example through access concepts and traceable refinement layers. Security techniques such as row-level security are concrete tools within governance.
How smiit works with it
smiit anchors data governance pragmatically in data platforms without making it an end in itself. Definitions, responsibilities, quality rules and permissions are designed so that they are followed in everyday work and build trust in the data.
Common mistakes & misconceptions
- Data governance is not just privacy or compliance; it covers roles, standards, data quality and responsibilities across the entire data lifecycle.
- Many treat governance as a one-off project, but it is an ongoing process that must evolve with the organization and the data landscape.
- A common error is to see governance as IT's job alone. Successful governance needs clear business data owners and data stewards.
Frequently asked questions
What is the difference between data governance and data strategy?
The data strategy defines which goals should be achieved with data. Data governance is the operational rulebook of roles, standards and processes that reliably implements those goals.
Who is responsible for data governance in a company?
Typical roles are data owner (responsible for a data domain) and data steward (looks after quality and maintenance). Overall responsibility usually lies with management or a designated data owner.
Is data governance the same as data protection?
No. Data protection, for example under the GDPR, is an important part of governance, but governance also covers data quality, definitions, responsibilities and access rules.
Does data governance only become relevant above a certain company size?
No. Even a handful of reports with inconsistent metric definitions lead to misunderstandings. In SMEs a lean governance with clear definitions, named owners and simple access rules is often enough, rather than a heavy rulebook.
How do you start with data governance pragmatically?
It makes sense to start small: define the most important metrics consistently, name owners for the central data domains and clarify access rights. Governance then grows with the data requirements instead of introducing every rule at once from the start.
Related terms
Sources & further reading
Want to put this topic to work in your company?
Updated · Back to the glossary