The migration vendor promised a clean single source within four weeks. Instead they applied loose matching rules that treated 'Acme Pty Ltd' and 'Acme Holdings' as separate suppliers even when bank details matched exactly.
Demand planning models now train on halved purchase volumes for dozens of critical parts. The AI flags stockouts that never existed while real shortages go unpredicted because the true history sits split across two records.
No one on the steering committee asked for a post-cutover duplicate audit. The project closed green because the dashboard showed 98 percent record migration completion, not because any downstream system could actually use the data.
Finance teams still manually reconcile the duplicates every month before running cash flow projections. The automation budget went to the migration, not to fixing the broken matching logic that created the mess.
Procurement now negotiates with phantom duplicates and loses volume discounts that the old system captured automatically. The commercial impact shows up in margin erosion rather than any single red flag in the data quality report.
Your next integration project will inherit the same split records unless the matching rules get rewritten with stricter identifiers. The vendor has already moved on to the next client and left the cleanup as an internal problem.