Leadership development as an operating system

The constraint

The leaders already had plenty of feedback. What they lacked was a mechanism that carried insight into the quarter. Peers kept building shadow trackers. Direct reports kept re-deriving their own priorities each week.

So I built the cycle as one system with a single test at the end: does the experience of working with this leader change, and does the change survive after the coach steps away?

What the diagnosis surfaced

I read the report for divergence: where a leader's own rating and their stakeholders' ratings separate, which groups see it, and which few behaviors sit underneath the gap. The gap identifies where intent and experience differ.

What made it usable

• Three lowest items named: one on expectations, one on priority stability, one on visible closure
• One strength selected deliberately to help carry the change
• One real gap, psychological safety, explicitly deferred: three priorities at once produce three partial results

This is the pivot the whole system depends on. Every commitment is written so a third party can tell whether it happened without asking the participant.

• Written so it can be observed

  Issue any priority change in writing within 24 hours and name what is being dropped to make room.

• Written so it can be observed

  Write a one-page role scorecard for each direct report (outcomes owned, decision rights held, criteria their work is judged against) and review it live with each person.

• Written so it can be observed

  Post closure in the same thread where the commitment was made, on or before the date; if the date will slip, post a revised date before the original passes.

• Written so it can be observed

  Send peer function leads a monthly written recap: what closed, what slipped, what changed.

I use the five team conditions as diagnostic lenses, asking leaders to test whether their teams actually experience the conditions they believe they create.

Select a condition to see the question I put to the room.

Holding the tension

When a senior leader says “people here can speak up, we're already an open culture,” I treat it as a hypothesis to test against the team's experience. Then I ask the room the question above and stop talking for eight to ten seconds.

If the answer is hard to retrieve, that is data. The session closes with each leader reading their committed behavior out loud to a partner. The partner tests one thing: could they recognize the behavior when it happens? Anything too vague gets rewritten.

Facilitation excerpt uses fictional scenarios and participant details. The original program deck remains Company Confidential.

What the session produced

The Project Aristotle senior-leader workshop is measured through documented follow-through after the session.

• 20 participants
• 19 documented a behavioral commitment
• 16 completed the 30-day follow-up
• 14 could point to a concrete application of the behavior in their work

Counts come from the session record and 30-day participant follow-up. They reflect self-reported follow-through.

Select a step to see what it produces.

1. 01Written role scorecardsOutcomes owned, decision rights held, and the criteria the work is judged against, signed and on file. Two direct reports were found to be carrying work neither believed they owned.
2. 02Priority restatement
3. 03Explicit tradeoffs
4. 04Visible commitment closure
5. 05Evidence where the work occurs

The checkpoints did real work

At 30 days the plan was at risk. Several role scorecards were still unwritten, and the missing ones belonged to the regions in the middle of a structure change, deferred on the reasoning that the roles might move.

The adjustment was to write them against the current role, dated, marked provisional. Clarity work gets postponed precisely when conditions are unclear, which is when it matters most. By 60 days the practices were holding without prompting and support moved from weekly to biweekly.

The business verifies the behavior, keeping accountability with the operating line.

The participant's leader already held a regular business review. Clarity measures and visible commitment closure became part of that conversation, keeping accountability with the business line.

Practice capacity did the work here. The technology just made repetition cheap.

Before leaders used new scorecards, priority-reset language, or tradeoff conversations with their teams, they could rehearse against AI personas and digital voice tools tuned to the business context. The practice was low stakes, repeatable, and available on demand.

What this produces is lateral accountability across functions.

I embedded the behavior into tools leaders already used, with automated prompts to document a cross-functional tradeoff or close a commitment where the work was already happening.

• Standalone L&D task
• → embedded operating habit
• → automatic evidence trail
• → less enforcement required

• Psychological Safety, cohort average of 12

  9.8→10.4+0.6

• Meaningful Work, cohort average of 12

  10.4→10.7+0.3

• Dependability, cohort average of 12

  10.1→10.9+0.8

• Structure and Clarity, cohort average of 12

  9.6→10.9+1.3

• Impact, cohort average of 12

  10.5→10.8+0.3

• Core Values, cohort average of 12

  10.8→11.0+0.2

Same 15 leaders, same instrument, same scoring method. Six dimensions, three behavioral items each, rated 1 to 4 for impact and summed to a 3 to 12 dimension score; self ratings are excluded from the cohort figures. Approximately 90 percent of raters were unchanged, and replacement raters were matched by relationship type. The figures report observed movement on the repeated instrument across the matched cohort.

The operating evidence

• Peers stopped maintaining shadow trackers for Field Operations commitments
• Two peer leads named the monthly recap as the reason they stopped duplicating tracking
• Clarity practices survived a regional structure change in month five
• Closure rate is reported in the participant's own monthly business review
• Coaching support ended; the next contact was the reassessment

One direct report still rates clarity at 2, citing decision rights on vendor issues specifically. That is carried forward rather than averaged away.