Every maturity tool ultimately runs on self-reported answers. Why is your assessment data any more trustworthy than a survey my team fills in optimistically?

The full executive answer

Because we treat self-reporting as a data-quality problem and engineered against it, rather than pretending questionnaires are ground truth. The Evidence Map classifies every assessment dimension by how well it is corroborated: backed by multiple independent sources, supported, merely self-reported, or unevidenced. That classification is not a footnote — it mechanically moves the output. When telemetry or leadership signals contradict a flattering self-score, the confidence band on the verdict is widened downward on the server and the conflict is stamped visibly on the output; broad corroboration lifts the floor. The model is never asked to judge its own grounding — our code computes it.

So an optimistic survey does not produce a confident, flattering verdict — it produces a hedged verdict that says, in effect, "this organisation’s self-image is unverified in these areas", which is itself one of the most valuable findings a board can receive. In the demo I can show the same organisation with and without corroborating evidence, and you will watch the confidence band move.

And uniquely, the platform then submits its own judgments to reality: forecasts are logged as scoreable predictions and graded against later real assessments — predicted versus actual, error and within-band — building a public calibration record over time. A survey cannot be wrong; our forecasts can be, measurably, and that is what makes the whole system trustworthy. Pre-launch caveat stated plainly: the calibration ledger starts near-empty and only reality can fill it.