Description: Dex’s current duplicate detection is intentionally conservative to avoid incorrect merges, but this causes certain clear duplicates, such as first.last patterns matching company domains, to be missed. Users have requested a more powerful merge flow that can catch these nuanced cases. Proposed enhancements: Add an “Advanced Merge Review” mode with deeper matching logic Allow users to review and approve each suggested merge individually Expand detection rules to account for patterns like matching email formats, company domains, or slight name variations Maintain strict safety checks to prevent incorrect automatic merges This would help Dex surface more potential duplicates while still giving users full control, improving data hygiene without increasing merge risk.
