On paper, they are exemplary. The psychometric profile is clean, the 360 is glowing, the engagement score for their team sits comfortably above the org average. And yet, six months later, they are the resignation nobody modelled — the one that lands as a shock precisely because every instrument you own said the opposite.

The number is not the mechanism

Quantitative instruments are very good at one thing: telling you how much, and how it compares. They are structurally incapable of telling you why, or through what mechanism. A high score is a summary — and a summary is, by definition, the removal of the very texture a retention decision depends on.

This is not a criticism of the data. It is a description of its shape. When leadership treats a strong scorecard as reassurance, the score stops being a measurement and starts being a place to stop looking.

The strongest scorecard in the room is often the one that most deserves a second, closer read.

Three gaps we read for

In the field, we are not looking to contradict the numbers. We are looking for the distance between what the data reports and what the person actually experiences, week to week. Three gaps recur:

  • Effort vs. discretionary effort. The work still ships. What has quietly stopped is the extra — the initiative, the mentoring, the unasked-for improvement. Engagement surveys rarely separate the two.
  • Stated vs. enacted priorities. What a leader says matters and what their calendar, their language and their attention actually protect are often different documents.
  • Recognition landing vs. recognition given. Organizations track whether recognition is offered. They almost never track whether it is received as recognition — or as noise.

Why the highest scorers are exposed

The mechanism is almost banal once you see it. High performers are reliable, so they are asked more. Because they rarely complain, the asking is invisible. Because the score is high, no one investigates. The very competence that produces the strong number is what conceals the load underneath it.

By the time the signal reaches a survey, the decision has usually already been made privately, months earlier. The survey measures the aftermath, not the turning point.

What to do with a "too good" score

Not more surveying. A number cannot be resolved by another number. The move is qualitative: a structured, confidential read of the mechanism — how the role is actually experienced, where discretionary effort is going, what the person has stopped saying out loud.

None of that appears in a bar chart. All of it appears in a room, if the room is built for it. Reading the gap early is the difference between a retention conversation and an exit interview.