Data Quality

The Complete Data Profiling Checklist for MDM Projects

A step-by-step checklist for conducting thorough data profiling before any MDM implementation — ensuring no surprises during migration and go-live.

March 9, 2025

6 min read

Why You Need This Checklist

Data profiling is the most underestimated phase of MDM projects. Teams that skip or rush profiling consistently face cost overruns, timeline delays, and data quality issues post-go-live. This checklist ensures nothing is missed.

Phase 1: Source System Inventory

[ ] Identify all source systems containing master data

[ ] Document system owners and data stewards per source

[ ] Catalog data formats and access methods (API, DB, file)

[ ] Note data refresh frequencies and latency

[ ] Map current data flows between systems

[ ] Document known data quality issues per source

Phase 2: Structure Profiling

[ ] Catalog all tables/entities containing master data

[ ] Document column names, data types, and lengths

[ ] Identify primary keys and unique constraints

[ ] Map foreign key relationships

[ ] Count records per table with growth trends

[ ] Identify unused or deprecated columns

[ ] Check for schema inconsistencies across environments

Phase 3: Content Profiling

[ ] Calculate null rates for every column

[ ] Generate value frequency distributions

[ ] Identify min/max/average for numeric fields

[ ] Detect patterns in string fields (formats, prefixes)

[ ] Check for default/placeholder values

[ ] Identify encoding and character set issues

[ ] Profile free-text fields for quality and consistency

Phase 4: Quality Assessment

Completeness

[ ] Define mandatory fields per entity type

[ ] Calculate completeness scores per source

[ ] Identify records missing critical attributes

[ ] Estimate remediation effort for incomplete data

Accuracy

[ ] Validate addresses against postal databases

[ ] Verify identifiers (GTIN, DUNS, tax ID)

[ ] Cross-reference with authoritative sources

[ ] Sample manual verification for critical fields

Consistency

[ ] Compare same entities across source systems

[ ] Identify conflicting values for shared attributes

[ ] Check referential integrity across tables

[ ] Validate hierarchical relationships

Uniqueness

[ ] Run duplicate detection on key matching fields

[ ] Calculate duplicate rates per source

[ ] Identify cross-source duplicates

[ ] Estimate merge complexity

Phase 5: Relationship Profiling

[ ] Map entity relationships across sources

[ ] Identify orphaned records (missing parents)

[ ] Detect circular references

[ ] Profile hierarchy depth and breadth

[ ] Document relationship cardinality

Phase 6: Volume & Performance

[ ] Total record counts per entity type

[ ] Data growth projections (1, 3, 5 years)

[ ] Historical data volumes and retention needs

[ ] Peak transaction/update volumes

[ ] Storage requirements estimation

Phase 7: Profiling Report

[ ] Executive summary with key findings

[ ] Data quality scorecards per source

[ ] Recommended cleansing rules and priorities

[ ] Data model design recommendations

[ ] Migration effort estimates

[ ] Risk register for go-live readiness

[ ] Detailed appendices with profiling statistics

eTron Digital's Profiling Methodology

This checklist is based on our proven profiling methodology used across 10+ MDM implementations. We deliver comprehensive profiling reports that eliminate migration surprises and accelerate implementation timelines.