Skip to content
SF CTA Study Hub

Data: Quick Reference

Fast-track refresher for Domain 3: Data Architecture and Management. Read this if you have 2 hours before a mock board. For deep dives, see Domain 3 full content.

What the Board Expects

The review board evaluates your ability to design data architectures that balance performance, scalability, and maintainability. You must demonstrate:

  • Deliberate relationship type selection with trade-off articulation
  • LDV awareness — indexing strategy, skew mitigation, archival planning
  • Migration phasing with tool selection rationale and rollback planning
  • Data quality and governance baked into the design, not bolted on

Quick Reference Files

FileCoversKey Artifacts
Data Modeling Quick RefRelationship types, limits, ERD patterns, anti-patternsRelationship decision table, ERD templates, gotcha callouts
LDV Quick RefThresholds, indexing, selectivity, skew, archivalLimits cheat sheet, skew decision tree, selectivity thresholds
Migration Quick RefPhases, tool selection, sequencing, cutover strategiesMigration phase diagram, tool comparison matrix, load order

Domain 3 Board Presentation Checklist

Before presenting, confirm you can answer these:

  • Why did you choose master-detail vs lookup for each relationship?
  • Which objects will exceed 1M records? What is your index strategy?
  • Where is data skew likely? How do you mitigate it?
  • What is your migration sequence? Why that order?
  • What cutover strategy did you pick and why?
  • What is your archival and retention policy?
  • How do you handle data quality at entry, integration, and ongoing?
  • What compliance requirements affect your data design (GDPR, residency)?

The 5-Minute Framework

When you see a data scenario, work through this sequence:

graph LR
    A[1. Identify<br/>entities] --> B[2. Choose<br/>relationships]
    B --> C[3. Assess<br/>volume]
    C --> D[4. Plan<br/>indexes & archival]
    D --> E[5. Design<br/>migration]
    E --> F[6. Define<br/>governance]
  1. Identify entities — Standard objects first, custom only with justification
  2. Choose relationships — MD vs lookup based on sharing, roll-ups, independence
  3. Assess volume — Which objects hit LDV? Growth projections?
  4. Plan indexes and archival — Custom indexes, skinny tables, Big Objects, external storage
  5. Design migration — Sequence, tools, cutover strategy, trial runs
  6. Define governance — Quality controls, stewardship, retention, compliance

Deep Dive References

Sources