When the factors update, your baseline should not quietly change.

The controls that make a reported emissions figure reproducible: the factor version pinned to the period that used it, the period locked when it closes, the calculation recorded rather than recomputed, and a trace from the figure back to its source. It is the layer beneath reporting, not a reporting feature.

Storing which conversion factor set, at which publisher revision, priced each figure, so the figure keeps the factors in force for its period. Without it, republishing the annual factors silently restates prior years, and the first person to notice is usually an auditor comparing this year’s pack against last year’s.

It closes the reporting period so figures inside it cannot change without an explicit, recorded restatement. Late data arriving against a closed period is held for review rather than absorbed silently, which is what makes a prior-year comparison trustworthy.

The calculation fails rather than borrowing another year’s factors. An interim set carries a label that travels into every trace and citation produced from it, so a figure priced with provisional factors is identifiable as such wherever it appears.

A dashboard presents figures; governance determines whether they hold up when questioned. Most carbon software handles one scope at a time and leaves the joining, the versioning and the evidence to you. The distinction only becomes visible under assurance, which is also the worst time to discover it.

It moves the work earlier. Recording provenance as data arrives takes longer than importing it unchecked, and it removes the reconstruction exercise that otherwise happens under deadline when somebody asks how a number was produced.