Data governance is an organization's set of principles, policies, and processes to ensure data availability, quality, and security for data consumers. These defined governance "laws" apply specifically to data consumers (both human and machine) because they are the primary beneficiaries of successful data governance.
A data governance office or data stewards usually execute data governance. They will compile rules and processes that govern how the company handles data and makes that information available to anyone who needs it. These individuals will monitor and evaluate these rules, fixing them when necessary. Data governance can then be broken into three components: data quality, master and reference data management, and metadata management.
Examples of these rules could be a time limit for how long personal data should be stored or a rule defining metrics the company uses to determine acceptable data quality.
You want your data governance program to succeed for two reasons: data availability and trust in data. Data governance can help accessibility by defining master, reference, and metadata management rules. This will aid search in the data catalog for terms, data owners, data sources, and other metadata. It will also help understand relationships between data and business terms (and their meaning) for metadata analysis.
Data governance will improve your organization's trust in data by providing data authenticity (understanding the origin and quality of a data set) and data quality transparency (understanding the state of data in any available source).