Skip to main content

15.9 Hands-on Governance Labs

These labs are intentionally scenario-driven.

The objective is not to memorize catalog buttons. The objective is to reason like a governance architect.

Lab 1 — Orphaned critical asset​

Situation​

A Tier 1 Finance asset has no owner.

Task​

Design the remediation workflow.

Expected reasoning​

  • identify domain,
  • assign accountable owner,
  • assign steward,
  • classify criticality,
  • review consumers,
  • update catalog metadata,
  • verify quality and lineage.

Lab 2 — Duplicate Customer datasets​

Situation​

Five teams have built different Customer datasets.

Task​

Determine whether they should be consolidated.

Analyze​

  • grain,
  • business definition,
  • source,
  • consumer requirements,
  • quality,
  • freshness,
  • ownership.

Do not consolidate simply because the names are similar.

Lab 3 — Glossary conflict​

Situation​

Finance and Sales use different definitions of Revenue.

Task​

Create an enterprise decision process.

Required output​

  • competing definitions,
  • business owner,
  • policy reference,
  • approved terms,
  • migration impact.

Lab 4 — Data product proposal​

Create a Procurement Spend data product.

Define:

  • purpose,
  • consumers,
  • grain,
  • quality,
  • owner,
  • steward,
  • freshness,
  • classification,
  • access,
  • lifecycle.

Lab 5 — Quality incident​

Situation​

Daily sales data arrived successfully but contains duplicate order lines.

Task​

Explain why a successful pipeline is not evidence of quality.

Design:

  • detection,
  • severity,
  • owner,
  • consumer impact,
  • remediation,
  • validation.

Lab 6 — Breaking change​

Situation​

A product changes its customer key.

Task​

Build an impact-management plan.

Include:

  • lineage,
  • downstream models,
  • analytics,
  • data products,
  • access,
  • communication,
  • compatibility.

Lab 7 — Catalog enrichment​

Situation​

A critical dataset is searchable but poorly documented.

Task​

Define the minimum enrichment required before certification.

Lab 8 — Governance at scale​

Situation​

The organization has 10,000 assets and only 12 stewards.

Task​

Design a scalable governance model.

The correct answer should rely on:

  • criticality,
  • automation,
  • domain ownership,
  • policy-as-standard,
  • targeted stewardship.

Do not propose manual review of every object.

Lab 9 — Product retirement​

Situation​

A data product has 30 active consumers and a better successor exists.

Task​

Design a safe retirement plan.

Include:

  • consumer discovery,
  • migration,
  • deprecation notice,
  • monitoring,
  • retirement criteria.

Lab 10 — Architect review​

Present your proposed governance architecture in ten minutes.

You must answer:

  1. Who owns data?
  2. Who defines meaning?
  3. Who controls access?
  4. Who monitors quality?
  5. Who publishes products?
  6. How do consumers discover data?
  7. How are changes assessed?
  8. How are incidents escalated?
  9. How is governance measured?
  10. How does the model scale?

If any answer is "the central team handles it," explain why that does not become a bottleneck.

Deliverable​

For each lab, produce:

  • context,
  • decision,
  • alternatives,
  • architecture,
  • controls,
  • risks,
  • metrics.

This is the same structure you should use in enterprise architecture interviews.