use cases · Leadership measurement
You're measuring whether people like their leaders — not whether their leaders are right
You're measuring whether people like their leaders — not whether their leaders are right. The right signal is measurable.
For who
What it finds
What you get
Binding constraint
The situation
The exec team wants to be 'data-driven about leadership,' but the instruments are 360s, engagement scores, and leadership-competency vibes. Promotions and succession lean on who's well-liked and tells a good story. The plan: another 360 cycle, a new competency model, more leadership training — none of which is graded against anything real.
How the walkthrough goes
- 01customer-situation
You want to be data-driven about leadership — so you run another 360.
360s, engagement scores, a competency model. Promotions and succession lean on who's well-liked and tells a good story.
- 02problem-cost
You're putting your best-liked people in charge of more.
If likeability isn't judgment, you're compounding the wrong bet — promoting and succession-planning on a signal that doesn't predict outcomes.
- 03insight
You're measuring perception, not judgment.
Leadership quality is a capability you can grade against reality — do they actually know their teams (forecast accuracy) and hold honest uncertainty (calibration)? Your best-liked leaders are not your best-calibrated ones. And these capabilities are learnable.
- 04desired-outcome
Promote and develop on what predicts real outcomes.
Grade leaders against reality — judgment, calibration, resource-match — and coach the gaps, because they improve.
- 05product-path
Performix plus the leadership battery grade against reality.
CAMS binding-constraint at the team level, with forecast-strength and alignment constructs from the measurement backbone.
- 06proof
Likeability doesn't predict outcomes. Calibration does — and it's learnable.
In the data, 360/likeability doesn't separate the leaders whose teams perform; forecast accuracy and calibration do — and they improve with practice.
- 07risk-reversal
A toolkit, not a scorecard to fear.
Every dimension is graded against something real (variance explained, forecast-vs-actual) — and every one is learnable. It's how leaders get better, not a ranking that punishes them.
- 08next-step
Measure the signal that predicts — before the next promotion.
One read on which leaders actually know their teams and call it straight, before you put someone in charge of more.
Grounded in the research
- — Measuring Leadership program (docs/strategy/MEASURING-LEADERSHIP-PROGRAM.md) — grade leaders against reality, not opinion: resource optimization · alignment · NAV · forecast strength (accuracy + calibration) · CAMS
- — Tetlock (Superforecasting / EPJ) — forecast accuracy and calibration are measurable and trainable; confidence ≠ accuracy
- — Kahneman, Sibony & Sunstein (Noise) — judgment quality is measurable and improvable; perception-based ratings are noisy
- — 360-degree-feedback validity literature — 360s capture perception and weakly predict objective performance
Walkthrough data is composite and clearly labeled — shaped from the research to show the real shape of the finding, not a named client.
Promote, develop, and succession-plan leaders on capabilities that predict real outcomes (judgment, calibration, resource-match) instead of likeability — the decision-error avoided is putting well-liked, poorly-calibrated leaders in charge of more.