Board Reporting
Comparability gating
The rules that withhold a comparison when the two assessments were not produced under the same methodology, account or readable snapshot.
The gates
| Condition | Withheld |
|---|---|
| The scoring fingerprint changed | Composite and domain score comparison. Control-level movement continues |
| The mapping version changed | Materiality band movement |
| A different account fingerprint | The comparison, outright |
| Snapshot missing, unreadable, or of an unknown version | The comparison, with the reason stated |
Why score comparison is gated but control movement is not
Control identifiers are stable across engine versions. A control that was failing and is now passing is the same fact regardless of how the composite was weighted.
Scores are not stable across a methodology change, and comparing them would present a changed threshold as a changed environment.
Why a different account is refused rather than caveated
Because there is no meaningful partial comparison between two accounts. Offering one with a warning invites it to be read.
First assessment versus unreadable predecessor
These are worded differently. A pack that presented an unreadable predecessor as a first assessment would be concealing a failure.
What it does not mean
A withheld comparison is not a finding. It says the two results cannot be placed on the same axis.