Skip to main content

Glossary

Glossary

This glossary defines key terms and concepts used throughout the Qarion platform to help you understand the data governance ecosystem.

A​

Access Request​

A formal request from a user to gain access to a data product or source system role. Requests go through an approval workflow and are tracked for audit purposes.

AI Copilot​

Qarion's built-in AI assistant that helps users discover data through natural-language questions and auto-generates documentation for data products. The Copilot provides source-attributed responses to guide users without requiring SQL or filter syntax.

AI System​

A data product type representing a machine learning model, AI pipeline, or automated decision-making system. AI Systems support risk classification aligned with the EU AI Act and can be monitored for data drift, model drift, and performance degradation.

Action Item​

A task created during a governance meeting, assigned to a specific person, and tracked to completion.

Alert​

A notification generated when a quality check fails or a threshold is breached. Alerts appear in the Alerts Center and can be linked to issues.

Approver​

A user with permission to approve or reject access requests, typically a product steward, owner, or space admin.

C​

Contract​

A formal agreement defining service levels (SLAs) between data producers and consumers. Contracts specify expectations for data quality, freshness, and availability.

Custodian​

A governance role responsible for technical maintenance of a data product, including schema updates and quality check configuration.

D​

Dashboard​

The main landing page for a space, showing activity feeds, key metrics, and quick access to common actions.

Data Architect​

A governance role responsible for designing data models and defining architectural standards for a data product, ensuring structural consistency across the organization.

Drift Detection​

Continuous monitoring that identifies when data distributions, model outputs, or relationships shift from their expected baselines. Qarion supports data drift, model drift, concept drift, and performance degradation monitoring for AI systems through the Quality Engine.

Data Catalog​

The central repository where all data products in a space are registered, documented, and discoverable.

Data Lineage​

A visual representation of how data flows from source systems through transformations to downstream products and consumers.

Data Product​

A registered data asset in the catalog, such as a table, dataset, dashboard, or API. Products have metadata, schema information, quality checks, and governance assignments.

F​

Field​

A column or attribute within a data product's schema. Fields have names, types, descriptions, and structural indicators (PK, FK, PII, etc.).

G​

The platform-wide search accessible via ⌘K (Cmd+K) that finds products, issues, meetings, source systems, and users.

Governance Meeting​

A scheduled meeting for reviewing data governance topics, with tracked agendas, notes, and action items.

I​

Issue​

A tracked problem or task related to data quality, typically managed on the Kanban board. Issues have priorities, statuses, and can be linked to products and alerts.

L​

Lineage Graph​

The visual node-and-edge diagram showing upstream and downstream relationships between data products.

O​

Owner​

The highest governance role for a data product, responsible for strategic decisions and approving major changes.

P​

PII (Personally Identifiable Information)​

Data that can identify an individual. Fields marked as PII require special handling and access controls.

Primary Key (PK)​

A field or combination of fields that uniquely identifies each record in a dataset.

Product​

See Data Product.

Q​

Quality Check​

An automated validation rule that runs against a data product to verify data quality. Checks have types (freshness, volume, custom SQL) and execution history.

Quality Score​

An aggregate metric indicating the overall health of a data product based on its quality check results.

R​

Risk Classification​

A label assigned to AI System products indicating their risk level under the EU AI Act framework. Levels include unacceptable, high, limited, minimal, and none. Changes to risk classification are tracked in the product's audit history.

Role​

A named permission set that can be assigned to users. Roles exist at the space level and within source systems.

S​

Space​

An isolated workspace containing data products, quality checks, contracts, and governance processes. Spaces typically represent teams, departments, or business domains.

Source System​

A connection to an external data platform (Snowflake, BigQuery, etc.) where actual data resides.

Steward​

A governance role responsible for day-to-day curation of a data product, including documentation, access approvals, and quality monitoring.

Sync Job​

A background process that imports or updates metadata from a connected source system.

T​

Tag​

A label applied to data products for categorization and filtering. Tags are organized in taxonomies.

Taxonomy​

A hierarchical organization of tags used to categorize data products.

U​

Upstream​

Data products that feed into or are dependencies of a given product in the lineage graph.

Downstream​

Data products or consumers that depend on a given product in the lineage graph.