Data Office
The Data Office is delegated by the Data Owner to manage data access and compliance in OSDU at Equinor. It governs who can access data, performs and approves Legal Risk Assessments, and ensures data entering the platform meets governance requirements.
Data Owner vs Data Producer vs owner groups
These three terms are easy to confuse but mean different things:
- Data Owner — a governance role. The organisational unit in Equinor ultimately accountable for a data domain (e.g., the unit responsible for Master Data).
- Data Producer — an operational role that ingests and maintains data in ADME on behalf of the Data Owner. Before getting write access, they must be approved by the Data Office (e.g., SMDA is the Data Producer for Master Data).
data.***.owners— a technical ACL group on each record that grants write access. Despite the name, it is the Data Producer (not the Data Owner) who is a member of this group.
The Data Owner is accountable for access decisions; the Data Office operationally manages who can read and write data. The Data Producer only ingests data — they do not govern access.
How it works together
Every record in OSDU is tagged with both a viewer group and an owner group:
- Viewer group — who can read the record. Managed by the Data Office.
- Owner group — who can modify the record. The Data Office assigns the Data Producer to this group.
The Data Office governs access by country and asset through viewer groups, assigns Data Producer teams to owner groups for write access, and performs and approves Legal Risk Assessments across all data classifications.
Data Governance Forum
The Data Governance Forum is the central decision-making body for data governance on the ADME platform. It is led by the data leads from the relevant data offices and includes representatives from Subsurface (EPI, EPN), Drilling & Wells, and TDI.
The forum is responsible for:
- Defining and maintaining data governance policies, standards, and best practices across ADME
- Ensuring data management processes align with regulatory, security, and organisational requirements
- Resolving cross-functional data issues and approving exceptions to governance policies
- Providing strategic direction and oversight for data ownership and management roles
Contractual agreement
Before a data producer team is granted write access to the production environment, they must adhere to the guidelines established by the Data Governance Forum. These guidelines include a contractual agreement that must be approved by the relevant data office. This gate ensures that data entering production meets governance standards and that the responsible team has committed to maintaining data quality and compliance.
Data classification
Data in ADME is classified through the Legal Risk Assessment (LRA) tag attached to each individual record. Currently, the platform contains only Public and Internal data.
See Legal Tags (reference) for the LRA tag format and links to the LRA Data Inventory.
Roles and responsibilities
| Role | Responsibility |
|---|---|
| Data Governance Forum | Defines governance policies, resolves cross-functional data issues, provides strategic oversight |
| Data Owner | Formally accountable for a data domain; approves policies and LRAs, and delegates operational responsibilities to the Data Office |
| Data Office | Manages data access (viewer and owner groups) and performs and approves Legal Risk Assessments |
| Data Producer | Ingests and maintains data on behalf of the Data Owner; must get write access approved by the Data Office before ingesting into production |
| Data Consumer | Uses the data for analysis, applications, or reporting |
| OSDU Platform Team | Manages platform configuration, access control, and supports onboarding (infrastructure is managed by Microsoft as part of ADME) |
Further reading
- Data Ownership (Omnia) — detailed overview of data ownership roles and responsibilities at Equinor
- Data Office Groups (reference) — group names, naming patterns, and ACL assignment rules
- Getting Access to Data — how to request data access
- Ingesting Data — entitlements, legal tags, LRA, and technical assurance