Data Governance

Data Lineage

Track data flow from source to consumption:

Source Systems → Raw Layer → Staging → Marts → Reports
     │              │           │        │        │
     └──────────────┴───────────┴────────┴────────┘
                    Lineage Graph

dbt Lineage

dbt docs generate
dbt docs serve
# View DAG at http://localhost:8080

OpenLineage

from openlineage.client import OpenLineageClient

client = OpenLineageClient(url="http://marquez:5000")

# Emit lineage events
client.emit(run_event)

Data Catalog

Metadata to Track

Category	Examples
Technical	Schema, types, partitions
Business	Description, owner, SLA
Operational	Freshness, quality scores
Access	PII classification, roles

dbt Documentation

models:
  - name: dim_customer
    description: "Customer master data"
    meta:
      owner: "data-team"
      pii: true
      freshness_sla: "24h"
    columns:
      - name: email
        description: "Customer email"
        meta:
          pii: true
          masking: "hash"

Access Control

Column-Level Security

-- Snowflake
CREATE MASKING POLICY email_mask AS (val STRING)
RETURNS STRING ->
  CASE
    WHEN CURRENT_ROLE() IN ('ADMIN') THEN val
    ELSE '***@***.com'
  END;

ALTER TABLE customers MODIFY COLUMN email
SET MASKING POLICY email_mask;

Row-Level Security

-- Snowflake
CREATE ROW ACCESS POLICY region_policy AS (region VARCHAR)
RETURNS BOOLEAN ->
  region = CURRENT_ROLE();

Compliance

• GDPR - Right to deletion, data export
• CCPA - Consumer data rights
• SOC2 - Access logging, encryption
• HIPAA - PHI protection