Dry-Run Duplicate Removal Preview Keeping the Most Complete Record

How it works

1. 1.Run Manual Trigger
2. 2.Emit six fictional records, or one _empty_batch marker item if the list is empty
3. 3.Add _keyField and _importantFields to each item
4. 4.Score, group and assign keep/drop/invalid_key with reasons
5. 5.Split keep vs others with IF
6. 6.Append both branches with Merge
7. 7.Build one report sorted by original index

Use cases

• Review customer lists before real deletion
• Clean asset records while preserving the fullest copy

Requirements

• No credentials needed
• Edit key field and important fields in the Set node

Setup and testing in n8n

1. Import the JSON into n8n
2. Run manually and inspect the last node output
3. Confirm C-100 keeps index 1 and drops index 0
4. Confirm C-200 keeps index 2 and drops index 3 on a tie at score 3
5. Confirm index 4 with an empty key is invalid_key with score 3
6. Set records = [] in the sample node and run: expect one report with empty_batch=true, all counts 0 and reconciled=true

Service Value

Ready-made workflow template for automation delivery and service execution.