sfdmu-data-plans

Use semicolon-delimited field names like "Field1;Field2", not the v4 "$$Field1$Field2" syntax. Relationship traversals such as Parent.Field or GrandParent.Parent.Field are supported, and $$ composite columns remain valid only in CSV headers for source-record matching.

Only with explicit user approval and a concrete current reason, since deleteOldData wipes all existing records before inserting. On the 5.6.4+ floor, the historical traversal-matching bugs are fixed, so Upsert is the correct operation for relationship-traversal externalIds.

This is the still-live Bug 4: SFDMU cannot decompose a composite $$ value in a lookup reference column, for both self-referential and cross-object relationships, leaving lookups null. Use a simple single-field reference such as ParentGroup.Code instead.

Upsert and Update plans match on externalId without duplicating, but Insert without deleteOldData duplicates every row, and Insert with deleteOldData can be blocked by live records referencing design-time rows. Load individual records surgically instead, then commit the rows to the plan CSVs.

If the object should load, restore the header row; if the object genuinely has no data in a wired plan, set excluded: true to prevent a destructive wipe; if the whole plan is unwired and unmaintained, delete the plan rather than masking it with excluded flags.