How TACTIC handles insufficient evidence
A submission that does not support a sub-dimension is the normal case, not the broken one. What matters is whether the assessment says so.
The two questions are kept apart
A sub-dimension can score low because the venture is weak on it, or because the submission did not say. Those are different findings and a report that conflates them is worse than useless to a panel, because the reader cannot tell whether to reject the venture or ask it a question.
So TACTIC counts them separately. Every Evaluation and Deep Assessment report states how many of its sub-dimensions scored at the floor, how many the submission supplied nothing for, and how many were both, and it states each of the three even when it is zero. Screening and Preview do not break a venture into sub-dimensions at all, so there is nothing there to count. Those figures are computed from the scores themselves, never asserted by the narrative.
Why a low score and an absent one are not the same number
On Evaluation and Deep Assessment, a component scored above the floor whose evidence cannot be found in the submission is floored before the score is combined. The report still counts the two sets separately, so a reader can see where the assessment had something to work from and where it did not.
The narrative cannot contradict the count
A sentence claiming a different number of unevidenced sub-dimensions from the one the scores support fails the render in development and is reported to our error tracking in production. The tables are the record; the prose describes them.
What this means if you are the one being assessed
An “insufficient evidence” marker is an instruction, not a verdict. It names the thing your submission did not cover, across four weighted dimensions. Adding that material and running the assessment again is the intended path, and both runs are kept so you can compare them.
The passage behind a score, and where it is shown
On Evaluation and Deep Assessment, a sub-dimension scored against the documented components records the passage of your submission its score rests on, and the passage is checked against what you submitted. The report and your dashboard print that passage under the score once it has passed the check.
Where no passage is printed, the line says why: the submission supplied nothing on that sub-dimension, the passage cited was not found in the submission, or it was not checked. A citation that failed its own check is never printed as if it were evidence.
Preview, Screening and the Dossier score the submission as a whole rather than component by component, so they carry no passage to show.