Flatten Variable Key/Value Pairs into Long Label/Value Rows per Deal

How it works

1. 1.Manual Trigger starts execution with an empty item
2. 2.A Code node emits a fictional demo payload of two deals with fully different keys
3. 3.A Code node flattens each deal into one row per key/value pair (deal_id included) with dot-path flattening
4. 4.A Set node normalizes the four final columns
5. 5.An If node splits empty-valued rows from filled ones
6. 6.Two Set nodes tag rows as filled or empty_value
7. 7.A Merge node in append mode recombines both branches; a NoOp is the final checkpoint

Use cases

• Normalizing CRM deals with variable keys before reports or spreadsheets
• Flattening nested JSON payloads into a repeatable fixed-column shape
• Data quality checks by isolating empty fields

Requirements

• No credentials needed; fully local execution
• For real data, replace the demo payload Code node with CRM/HTTP output using schema deals: [...]

Setup and testing in n8n

1. Import the JSON into n8n and run the Manual Trigger
2. Verify 14 output rows (7 per deal, including the field_label=deal_id rows) with fixed columns and stage/notes/probability/region tagged empty_value
3. Try a payload with completely different keys; columns stay fixed and row count equals the flattened pair count
4. Test an empty payload or a deal with no pairs; the workflow must not fail

Service Value

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