Case Study
Problem
Surface-level questions and outcomes that depended on which interviewer ran the board
Intervention
Full-lifecycle behavioral selection process with bias controls, independent scoring, and a hard integrity screen
Deliverable
Presented to executive leadership, who approved it and directed implementation across the organization
At a senior military leadership academy, instructors are entrusted with developing experienced leaders. Selecting them was high-stakes, but our process had two failures. The questions explored candidates only at the surface, and outcomes depended on who happened to run the interview. The same candidate could succeed or fail based on which board they drew, and even the successful ones were never truly tested.
That was not a style difference among interviewers, nor a failure of effort or intent. It was a process failure, the kind most organizations never realize they have. I rebuilt the entire process to assess demonstrated behavior against a standard that belonged to the institution, not the individual.
I designed a full-lifecycle behavioral selection process, from application screening through post-board feedback, and presented it to executive leadership as a complete, defensible package. They approved it and directed its implementation across the organization, where it is now the standing selection process. What follows are the design decisions that made it work, and the reasoning behind each one.
Role
Architect and author; presented to executive leadership, who approved it for organization-wide implementation
Problem
Surface-level questions and outcomes that depended on the interviewer
Scope
Seven-stage lifecycle: screening, demonstration, structured interview, scoring, deliberation, feedback, records
Foundation
Behavioral interviewing (STAR+L), weighted competencies, structured scoring
D-01
Every interview question requires a specific past event told in a structured format: Situation, Task, Action, Result, and the Lesson that changed how the candidate leads now. The panel does not accept hypotheticals or leadership philosophy.
Why
Anyone can describe good leadership. Only past behavior predicts future behavior, and adding the Lesson element tests whether the candidate actually grew from the experience or just survived it.
D-02
Five competencies are scored and weighted, but a disqualifying score in integrity ends the candidacy regardless of performance everywhere else. Any panel member can raise the concern at any time.
Why
The role develops people. A brilliant instructor with a character deficiency is not a reduced asset. They are a liability multiplied across every person they teach. Weighting cannot express that. Only a veto can.
D-03
Every candidate leads a live discussion with the panel, and at a planned moment a deliberately junior panel member pushes back or plays the person in the room who knows more. This is briefed in advance and happens every time.
Why
The job requires influencing experienced people who will challenge you. Most interviews leave that test to chance. I made it a repeatable, observable event, and gave the junior member a real scoring lane so their perspective could not be dismissed.
D-04
Every panel member records their scores before any discussion begins, and scores do not change once discussion starts. Members then brief their assessments in junior-to-senior order, with the senior member speaking last.
Why
These are two separate bias controls. Independent scoring stops members from drifting toward the senior opinion. Junior-first briefing stops them from softening what they say out loud after a senior voice has framed the candidate. Each control protects something different, and both are required.
D-05
Norming runs in two layers. Leadership and all senior panel members periodically agree, out loud, on what strong, weak, and disqualifying answers look like. Then each individual board holds a pre-brief the day before it convenes.
Why
A board that convenes without calibration sets the standard on the spot, which is the root cause of inconsistency. The two layers make separate boards consistent with each other and each board consistent within itself, so a personnel change never quietly resets the standard.
D-06
The panel informs the decision, but the recommendation belongs to a single senior member, with a written basis sufficient to reconstruct the decision from the file a year later. Only the organization's director can override it.
Why
Consensus diffuses accountability and lets a strong personality sway a weak panel. One name on the recommendation gives the institution one person to hold responsible and one decision to review, which is what accountability actually means.
D-07
Feedback to non-selects is delivered by leadership only, framed against the standard in specific, developmental terms. The phrase "not a good fit" is banned from the process.
Why
"Not a good fit" tells a developing leader nothing they can act on and reads as personal judgment, which is how grievances begin. Specific feedback treats every candidate the way the process asks instructors to treat every student, and that development compounds across the whole organization.
I presented the process to executive leadership as a complete package: governance structure, competency framework, scoring rubrics, interview banks, and every worksheet needed to run a board without me in the room. They approved it and directed its implementation across the organization, where it is now the standing selection process. That was the point. A standard that depends on its author is just another version of the original problem.
The design thinking is mine and I am glad to walk through any of it in depth. The document itself is internal to the Marine Corps, which is exactly how I treat proprietary work.