Large Scale Identity Data Migration

Issues and Considerations

Why This Matters

  • Data migration is often more costly and time-consuming than expected.
  • Poor planning leads to project failure.
  • Without a proper audit:
    • System integrity is not assured.
    • Hidden issues may cause data loss or incorrect outcomes.

Data Audit Techniques

  • Sampling
  • Manual review
  • Automated tools

Prominent Issues in Data Migration

  • New biometric matches
  • Data inconsistencies
  • Technical implementation errors
  • Data reshaping challenges
  • Remediation gaps
  • Parallel operations difficulties

New Biometric Matches

  • New algorithms expose previously undetected issues:
    • Matches on previously unlinked individuals
    • False positives on prior candidates
    • Records with low biometric quality (“untemplate-able”)

Data Inconsistencies

  • Example issues from legacy systems:
    • Special characters in names
    • Inconsistent date formats
    • Daylight savings effects
    • Legacy business case support
    • ...

Technical Implementation Problems

  • Unique identifier mix-ups (biometric & biographic)
  • Invalid XML/JSON in the database
  • Error codes stored in place of actual data
  • ...

Data Reshaping Issues

  • Adapting old data to a new structure caused:
    • Data loss
    • Misrepresented relationships
    • Loss of visibility of important properties
    • ...

Data Remediation Challenges

  • Fixes applied without resolving root cause
  • Root cause resolved but no remediation applied
  • Overlapping issues obscure diagnosis
  • No clear “Data Migration Owner”:
    • Completion vs. Quality mindset conflict

Parallel Operations Risks

  • Hard to keep both systems consistent
  • Syncing processes often lag
  • Business decisions may be made with outdated data

Conclusion & Recommendations

  • Establish Migration team early
  • Plan and audit all migration steps
  • Use detailed data mapping specs
  • Conduct data landscaping upfront
    • Biometric Quality important aspect
  • Manage outputs across versions
  • Prioritize parallel operations support
  • Use automated reconciliation tools to ensure:
    • Traceability
    • Operational assurance
    • Targeted remediation
    • Confidence in switchover

Copy of Copy of BQAT Gates Presentation

By Ted Dunstone

Copy of Copy of BQAT Gates Presentation

  • 7