NORTH is not a vibe check. Every archetype, readiness band and maturity score is tested against synthetic organisations with known ground truth. Here is how we validate the signal.
A leadership diagnostic only earns trust when its errors are visible and bounded. We publish the thresholds we test against, the synthetic data we generate, and the cases where the classifier still gets it wrong.
The result is not perfection. It is a known error profile: 85% archetype accuracy, no drift on stable trajectories, 89% misalignment detection, and an execution maturity score that is mechanically tied to the same dimensions that produce the archetype.
Every release of the SIGNAL classifier must pass these thresholds before it reaches a respondent.
Every archetype is classified correctly more than 72% of the time on held-out synthetic organisations. The classifier is tuned against deliberate edge cases, not just easy centroids.
When an organisation is generated on a stable Governance Steward trajectory, the archetype does not change between T0, T6 and T12. Stability means stability, not statistical noise.
We inject deliberate perturbations into synthetic dimension scores and verify that the diagnostic flags the contradiction or the archetype mismatch. Real signal, not hindsight.
The execution maturity score is derived from the same five-dimension composite used to generate the archetype, so the diagnostic cannot invent a maturity number that contradicts the signal.
We create 500 synthetic organisations with known archetypes, dimension profiles and trajectory types. Each organisation produces three records (T0, T6, T12), giving 1,500 test points.
For misalignment tests we deliberately shift one or two dimensions against the archetype fingerprint — for example, making a Governance Steward score low on Accountability. The test then checks whether the diagnostic notices.
The same code that classifies real respondents classifies the synthetic data. A separate contradiction detector checks whether the dimension pattern is internally consistent with the returned archetype.
If any archetype drops below 72% accuracy, stable trajectories drift, or misalignment detection falls below 75%, we tighten the dimension fingerprints and re-run the pipeline until the thresholds hold.
The classifier matches a respondent's dimension pattern against these fingerprints. The ranges are deliberately tight so that edge cases surface as contradictions, not confident misclassifications.
NORTH reads how you trade off values under pressure. It does not predict promotion, failure, or stock price.
The same leader can surface a different archetype in a different organisational moment. The score is a photograph, not a tattoo.
We report accuracy on synthetic data with known labels. In the wild, response quality, fatigue and cultural framing can move that number.
The Brief and the Coalition Map are inputs for leadership conversation. They are not instructions.
The 8 archetypes, 5 dimensions, 4 readiness bands, and the philosophy behind the NORTH diagnostic.