Aglet

Investigate Wrong Fields in an Import Mapping

Investigation should follow a source value until the destination field is observable. Use paired rows that expose names, positions, types, defaults, and empty values, then replay the mapping with a fixed revision. Keep source ambiguity separate from a mapping implementation defect.

Build a useful investigation brief

  1. Build labeled import rows

    Create fixtures with distinct values in every similar source column, plus empty, zero, malformed, and boundary values. Label the expected destination field and type before running them. Include the header versions and optional columns that were present during the affected import.

  2. Trace each mapping stage

    Capture raw headers and cells, parsed attributes, normalized values, mapping lookup, validation result, default application, stored fields, and rendered output. Check positional assumptions, case or whitespace handling, duplicate headers, dropped columns, and type conversion at every boundary.

  3. Compare revision and source

    Replay the same fixture with the affected and candidate mapping revisions, then compare a source format that imported correctly. Vary one header, column position, or value type at a time. If both source meanings remain plausible, request an owner decision instead of inferring intent.

What to carry forward

Produce a reproducible import brief with labeled rows, header and mapping traces, revision comparison, affected destination fields, and unresolved source meaning. Recommend the smallest mapping or contract change and a bounded repair cohort. Do not call a defaulted value correct merely because validation accepted it.

Keep the decision with the work.

Use a Work Item in Aglet to record the problem, the evidence you have, and the next decision. Add an owner and priority, then keep updates in the discussion so the next person can follow the reasoning.

Create an account See the product workflow