Columns missing from your BICC extract? Check isPopulate
Your model lists 502 columns but the file has 501. Some PVO columns are switched off by default — and rebuilding a job from the live job keeps them off.
You compare a BICC extract file against your column list and one column is simply absent — no error, no warning.
Why it happens
In a PVO's metadata (GET /biacm/rest/meta/datastores/{pvo}?expand=columns) each column has an
isPopulate flag. A small number of columns are isPopulate=false by default. If you build or
re-create a job by copying the existing job's column settings, those default-off columns stay off and never
appear in the extract.
isPopulate=true for every column you want. In one case this took a bank-account PVO from 501 to all 502
columns.How to spot it
- Compare the header of the extracted CSV with the columns your model expects, per run.
- Pad expected-but-missing columns as empty in the landing layer and log it, so downstream schemas stay stable while you investigate.
- Some “missing” columns are genuinely retired in Fusion (for example, a removed flexfield segment). The metadata tells you which case you're in.
Related: descriptive flexfields
DFF columns come and go as segments are configured per environment. Track them per instance, and treat a vanished DFF segment as a configuration change, not a pipeline failure.
See it on your own Fusion data
The free SQL editor queries tables and PVOs, and the catalog shows every PVO's columns and underlying SQL.
Request free access