AI Incentive Diagnostician
Describe a situation in your own words. The Diagnostician maps it across all six dimensions and returns the loop that's actually driving the behavior.
Run a diagnosis →
Describe a situation in your own words. The Diagnostician maps it across all six dimensions and returns the loop that's actually driving the behavior.
Run a diagnosis →A structured organizational read. Scores your incentive architecture across visible, hidden, perverse, missing, AI, and future incentives — and returns blind spots and interventions.
Take the assessment →Perverse incentives, cognitive bias, logical fallacy, mental models. The pattern library for spotting incentive failures in the wild.
Open the labs →How the framework reads Musk, Bezos, Jobs, and a roster of companies you've never heard of. Useful both as evidence and as a template for your own diagnosis.
Read the cases →A team. A meeting. A metric. A handoff. Big systems are hard to see; small systems make the dynamics obvious.
What's visibly rewarded? What's quietly rewarded? What's rewarded that shouldn't be? What's missing? What is AI optimizing for? What signal is forming?
A bad outcome is almost never about a bad actor. It's about a loop that makes the bad outcome the rational move. Name the loop.
You don't need to redesign comp. You need to change one signal in one loop and watch what moves. That's Stage 03 work.
A customer support org rolls out an AI assistant. CSAT rises. NPS quietly falls.
The AI is optimizing for the metric it can see: resolution time and first-contact closure. Customers feel handled, not heard. The human reward (close the ticket) and the machine objective (close the ticket faster) are aligned with each other and misaligned with loyalty. The fix is the objective, not the model.
A mid-size firm can't hire senior engineers despite top-quartile pay.
The future incentive — autonomy, optionality, equity in something they believe in — is forming faster than the comp market is repricing. Talent is leaving for those signals two to three years before salary surveys catch up. The visible incentive looks healthy. The future incentive is the leak.
A nonprofit board obsessively tracks fundraising; program outcomes drift.
Fundraising is measured weekly. Program impact is measured annually, qualitatively, by the people running the programs. Two compounding failures: a perverse incentive (raise more to raise more) and a missing one (no honest read on whether the money is producing the change the mission promises).
The Diagnostician is fast and situational — describe a moment, get a read. The Assessment is structured and organizational — it scores the whole architecture across all six dimensions and produces an executive summary you can hand to a board.
Yes. The same six dimensions apply to a career move, a co-founder split, or a parenting choice. The Diagnostician doesn't care about the size of the system — only about whether you can describe it clearly.
It's a thinking partner, not an oracle. Its job is to surface the dimensions you haven't named yet. The accuracy compounds the more honestly you describe what you're actually seeing.
If the diagnosis lands, the next move is usually one of three: a Workshop to redesign with peers, an Executive Course to build organizational fluency, or a conversation with the team about a tailored engagement.
Stage 03 is where individuals, teams, and organizations commit — to the Executive Course, an enterprise license, or a keynote that turns the read into a change.