You spot it by watching for evidence, not for resemblance. Most high-potential lists get built the same way every cycle: managers picture what a leader looks like, then scan the room for the person who matches that picture closest. The candidate who's quick to speak up, comfortable holding the floor, and visibly confident under scrutiny gets the nod. The person everyone quietly goes to before a hard decision, the one who defuses a blowup without needing credit for it, the operator whose team never seems to lose anyone, gets overlooked, not because they lack leadership behavior, but because they don't look like the mental template evaluators are unconsciously matching against.
If your high-potential list has looked almost identical for three review cycles running, that's not a coincidence. It's a measurement problem, and it starts in the brain of the person doing the evaluating.
Why This Is Harder Than It Looks
Every HR leader knows, in the abstract, that quiet people can lead. Nobody would argue otherwise if you asked them directly. The problem isn't belief, it's perception, and those are two different systems.
The people making HiPo calls aren't lying when they say they're looking for "leadership potential." They genuinely believe they're evaluating it. What they're actually doing is running a fast, automatic comparison between the candidate in front of them and a stored template of what a leader is supposed to look and sound like, built from every leader they've personally worked for, every executive they've watched present, every portrayal of leadership they've absorbed from television, business books, and conference keynotes. That template skews toward assertive, verbal, visible, and self-assured. It has almost nothing to do with what actually produces good outcomes on a team.
This is why the same names keep resurfacing. It's not favoritism in the usual sense. It's a perceptual shortcut that feels like judgment. The manager doing the nominating would tell you, honestly, that they're identifying leadership potential. They're identifying resemblance, and mistaking the speed and confidence of that match for insight.
The people who lose out aren't invisible. Their evidence is just harder to notice in a meeting, because it doesn't announce itself. It shows up in who gets pulled aside before a layoff conversation because they'll handle it with grace. It shows up in the team with unusually low attrition that nobody thinks to ask about, because nothing is on fire. It shows up in the person who ends an argument between two colleagues without anyone realizing a mediation just happened. None of that reads as "leadership" in a calibration meeting, because none of it looks like the template.
What's Actually Going On
There's a well-established line of research in organizational psychology called implicit leadership theory, first formalized by Robert Lord, Roseanne Foti, and Christy De Vader in 1984, and extended in Lord and Maher's broader information-processing model of how people perceive leadership. The finding is specific, and it's worth naming precisely rather than waving at "unconscious bias" in general.
People do not evaluate leadership by objectively observing behaviors and tracing them to outcomes. They compare the candidate against an internal, pre-loaded prototype of what a leader looks like, built up over a lifetime of exposure to leaders they've known and leaders they've watched from a distance. The comparison happens fast and beneath conscious awareness. The closer someone matches the prototype, the more "leadership potential" the evaluator perceives, and critically, that perception is largely independent of the candidate's actual leadership behavior.
Here's the mechanism made concrete. Picture two people in the same room, handling the same crisis: a furious client threatening to walk, mid-negotiation. One person is quiet and methodical. They ask two grounding questions, let the client vent without interrupting, and calmly walk them back from the ledge with a specific, credible plan. The other person is louder in the room. They jump in fast, project certainty, and talk over the tense silence that the first person was using deliberately. Both approaches work. The client stays. But in the debrief afterward, it's almost always the second person who gets remembered as "showing real leadership in that room." Not because their intervention did more, but because their behavior matched what leadership is supposed to sound like: fast, vocal, visibly in control. The first person's contribution gets filed under "handled it well," which is a very different mental bucket than "leadership potential," even though the underlying skill, staying calm under pressure and steering an outcome, is the same skill.
This is what makes prototype matching more dangerous than most bias conversations acknowledge. It doesn't feel like bias from the inside. It feels like recognition. The evaluator isn't ignoring evidence; they're not aware there's evidence they should be weighing at all, because the match already resolved the question for them before conscious evaluation started.
What to Do About It
You can't argue someone out of a perceptual shortcut by telling them it exists. Awareness of bias reduces it only marginally, and usually only for a few weeks after a training session. What actually interrupts prototype matching is changing what evaluators are required to look at, and when.
Start by separating the behaviors from the person before you talk about names. Before any calibration conversation, have nominators write down specific, dated incidents: not "she has great judgment" but "she was the one three people independently went to before the reorg announcement, and she talked each of them through it without escalating it to me." Not "he's a natural leader" but "his team has had zero voluntary turnover in eighteen months, in a department that's bleeding people everywhere else." Evidence forces evaluators to justify a nomination against something other than a gut read, and gut reads are exactly where the prototype does its work.
Build a criteria list before you look at candidates, not after. If you define what you're looking for only once names are already on the table, you'll unconsciously reverse-engineer criteria that fit the people the prototype already flagged. Define the behavioral markers first: who de-escalates conflict, who others seek out under pressure, who protects their team's stability without needing to announce it, and only then go looking for who exhibits them.
Finally, actively audit for the inverse signal. Ask your evaluators a different question than "who stands out as a leader": ask "whose team has outcomes that seem to require something we're not seeing," and "who do people quietly rely on that isn't getting mentioned." That question routes around the prototype entirely, because it asks about effects rather than appearance.
Signs You Need More Than This
If your organization has run through two or three review cycles and the HiPo list keeps regenerating the same handful of names, or the same personality type, in different job titles, that's a sign the problem isn't nomination technique, it's evaluator training. Individual managers correcting for their own bias one calibration meeting at a time won't hold, because the prototype resets by default the moment attention lapses.
This is the specific problem Dan Docherty, Braintrust's Chief Coaching Officer and author of NeuroCoaching, built a structured evaluation approach around: identifying leadership readiness from observed behavioral evidence against defined criteria, rather than from how someone comes across in a room. Braintrust's NeuroCoaching methodology is a direct antidote to prototype matching, because it removes the moment where a fast, confident-feeling impression gets to stand in for actual assessment.
The Real Test
The real test of a high-potential program isn't whether it can identify people who already look like leaders. Any manager can do that without help; the prototype does the work for them. The real test is whether the program can surface the people whose leadership doesn't announce itself, the ones who are already doing the work of leading without matching the picture anyone has in their head.
That's a harder problem than most HiPo programs are built to solve, but it's also where the actual bench strength is sitting, quietly, waiting to be noticed for what it already is.
Worth a conversation? Talk to Braintrust about building a leadership identification process that runs on evidence instead of resemblance.
