An organization designates a set of people as high potential. Some years later those people have advanced further than their peers, which appears to confirm the designation.
It confirms considerably less than it seems to, because the designation was not a neutral observation. It was an allocation decision, and the people who received it were then given the assignments, the exposure, and the advocacy that produce advancement.
The label does most of the work
Follow what happens to somebody after they are identified.
They are given more difficult work, which builds judgment. They are put in front of senior people, which builds standing. Their failures are read as development rather than as evidence, and somebody with authority is now invested in their success and will argue for them in rooms.
Any capable person receiving that package will outperform an equally capable person who did not. The subsequent divergence is real and it does not establish that the initial assessment was accurate.
Which means an organization cannot validate its identification process by looking at outcomes, because the process caused the outcomes. This is the central reason talent systems persist unexamined for decades.
Potential is relative to a situation
The word implies a general quality, and capability is considerably more specific than that.
Somebody who is excellent at holding a standard under pressure may be poor at building something from nothing. A person who is remarkable in a crisis may be unsuited to the patient work of a stable period. Judgment about people transfers less well than judgment about operations.
So the useful question is never whether somebody has potential. It is what specifically they might be capable of, and under what conditions, which is answerable and which the general label actively obscures.
An organization that has designated eight high-potential people has usually not asked what any of them are high potential for.
What the designation does to everybody else
Formal or not, the list becomes known, and it operates as a statement about everybody not on it.
Those people do not receive the assignments, so they do not build the judgment, so at the next review they present as less developed, which confirms their exclusion. The mechanism runs on its own and requires nobody to think badly of anybody.
It is particularly costly because capability appears at different ages. A person who was unremarkable at twenty-nine and formidable at thirty-eight is common, and a list constructed at twenty-nine and never revisited will have written them off permanently.
The practical correction is to revisit the list annually with a specific instruction: name somebody to remove and somebody to add. Forcing both directions prevents the list from becoming a permanent cohort, which is what every such list becomes if nobody intervenes.
Distribute the input, not just the assignments
If opportunity produces the outcome, then the allocation of opportunity is the decision that matters, and it is usually made by a small number of people drawing on a small pool of observation.
Two corrections are cheap. Ask supervisors who in their area would surprise you, which surfaces people whose reputation understates them. And check the list of who has received a consequential assignment in the last two years against the whole roster rather than against the existing cohort.
Neither of these requires a system. Both require somebody to be willing to discover that the current distribution is a product of proximity rather than judgment.
The person who was told
Being designated carries costs that are rarely acknowledged and that damage some of the people it was meant to help.
It creates an expectation of trajectory, which makes ordinary setbacks feel like failures of identity. It attracts scrutiny from peers. And it produces an incentive to appear ready rather than to be tested, which is precisely the opposite of what develops judgment.
Where somebody is told, the framing that helps is narrow and honest. We are going to give you harder things, and some of them will not go well, and that is what this is for. Not: we think you are going to be running this place.
The version that works
Stop asking who has potential and start asking who has been tested.
The first is a prediction dressed as an observation. The second is a fact about the organization’s own conduct, it is checkable, and acting on it produces the evidence that any assessment would have needed.
An organization that gives real decisions to a wide set of people and watches what happens does not need an identification process. It will have one, generated by observation of actual performance under actual consequence, which is the only kind that means anything.
Edited by Patrick J. Wolf, PhD