Data quality review
A steward-facing guide to the human review queues that clear data-quality gates.
Cleansing, mapping, and data review
Use this when stewards need to understand what to merge, dismiss, reopen, canonicalize, or send to Quick Fix.
Work the review queues
Each queue has a different decision type and export impact.
- Duplicate clusters are Pending, Merged, or Dismissed; merged clusters affect export survivorship.
- Value clusters create Quick Fix patches when a canonical value is applied.
- Anomaly scans emit advisory Medium issues and never auto-apply fixes.
Keep evidence clean
Review decisions must remain traceable after export.
- Merged duplicate clusters contribute survivor rationale to sign-off evidence.
- Version conflicts require reload before applying a stale decision.
- Canonical value choices become Quick Fix patches for affected records.
How to use this page
Each guide is a compact operating view. Use it to act, then continue into the related workflow page when you need the next step.
- 1ScanRead the section headings and confirm you are in the right guide for the role or moment.
- 2ActUse the bullets as the checklist for the extraction, review, approval, or handoff task.
- 3ContinueJump to a related guide when the work moves into upload, mapping, governance, or security.
Related guides
Use these when the workflow moves to the next handoff point.