Data Governance & Modeling

Technical strategies to ensure data trust, quality, and scalability. It moves beyond "moving data" to "managing data as a product".

Data Modeling Architectures

Pattern	Status	Context
Medallion Architecture	ADOPT	The standard Bronze/Silver/Gold layering. Simple, effective, and widely understood for Lakehouse setups.
Anchor Modeling	ASSESS	Lighter alternative to Data Vault for auditability and historization with less overhead.
Data Vault 2.0	ASSESS	Investigating for complex enterprise hubs where auditing and historical tracking are critical, despite the complexity overhead.
Dimensional vs. 3NF	ASSESS	Explicit guidance on when not to use Kimball and when normalized 3NF models are more appropriate.
Kimball (Star Schema)	ADOPT	The Gold standard for the presentation layer (Data Marts) to ensure fast BI performance.
Semantic / Metrics Layer	TRIAL	Unifying business metrics as a reusable layer to reduce metric drift across tools.
Event Modeling	ASSESS	Useful when building event-driven pipelines to keep behavioral flows explicit and aligned.

Tool	Status	Context
Great Expectations	ADOPT	Robust framework for testing data at ingestion, ensuring strict quality gates before processing.
Soda	TRIAL	Lightweight, SQL-native data quality monitoring for detecting anomalies in the warehouse.
Data Contracts	ASSESS	Defining data interfaces as code to prevent breaking downstream pipelines.
Datahub	ASSESS	Open-source metadata platform for end-to-end lineage and discovery.
SQLMesh	ASSESS	Monitoring this as a potential successor to dbt for more robust environment management and semantic understanding.
OpenLineage	ASSESS	Open standard for data lineage across orchestration, processing, and observability tools.
Amundsen	ASSESS	Open-source data catalog focused on discovery, ownership, and metadata search.