DEA Level 3 study plan

Map every data management and governance term in the exam outline into a one-page glossary with your own definitions

Draft a RACI-style ownership matrix for a fictional enterprise's core data domains

Compare three data modeling approaches (normalized relational, dimensional, and document-oriented) on a single worked example

Sketch a conceptual-to-logical model for a customer-order scenario and label every relationship's cardinality

Explain metadata management, data quality dimensions, and master data management in writing without notes

Trace how a governance policy flows from policy statement to enforceable control to monitored metric

Normalize a messy flat dataset to third normal form and then re-model it as a star schema

Identify and correct five common modeling defects (fan traps, chasm traps, missing grain statements)

Compare batch, micro-batch, and streaming ingestion patterns on latency, cost, and failure handling

Design an idempotent, restartable pipeline for a change-data-capture source

Write and run SQL that implements deduplication and late-arriving-record handling

Document data lineage for the pipeline from source to consumption layer

State how governance controls (retention, quality gates, ownership) attach to each pipeline stage

Implement slowly changing dimension type 2 logic and verify history correctness with test queries

Build a small dimensional warehouse from sample data and validate row counts at each layer

Distinguish descriptive, diagnostic, predictive, and prescriptive analytics with a worked example each

Compute and interpret core statistics (mean, median, variance, correlation) on a sample dataset and note when each misleads

Explain bias-variance tradeoff, overfitting, and train/validation/test splitting to a non-technical reader

Select appropriate evaluation metrics for classification and regression problems and justify each choice

Design a model documentation template covering features, assumptions, limitations, and monitoring

Assess how model feature choices depend on the underlying data model and its grain

Classify data into sensitivity tiers and assign handling rules to each tier

Map the principle of least privilege onto role-based and attribute-based access control designs

Outline the lifecycle of a data subject request from intake to verified deletion

Identify where encryption, masking, and tokenization each belong in a data architecture

List common privacy regulation obligations and match each to a concrete technical control

Write a one-page data strategy linking business objectives to measurable data initiatives

Build a phased program plan with dependencies, risks, and success metrics

Draft an incident response runbook for a suspected data breach

Practice translating a compliance requirement into an engineering backlog item

Take two full-length timed practice exams and log every missed question by domain

Re-study the three weakest sub-topics identified from practice results

Recite key definitions and frameworks from memory for each domain

Simulate exam-day pacing with a final timed mixed-domain set