The most common way a company loses two good people at once is to promote one of them. The best salesperson becomes the sales manager. Within a year the team is underperforming and the person who used to close everything is miserable, so they leave, and the six people who reported to them have already updated their profiles.
This pattern has been measured. Economists Alan Benson, Danielle Li and Kelly Shue examined 53,035 sales workers across 214 firms, 1,531 of whom were promoted into management, and found that doubling a worker's pre-promotion sales corresponded to a 7.5 percent decline in the sales performance of each of their new subordinates. Strong prior performance predicted promotion. It predicted managerial results in the wrong direction.
The practical problem this creates sits in your next talent review. Someone will put a name forward because the person had an exceptional year. Everyone will nod, because the evidence is right there in the numbers. And nobody in the room will be able to say cleanly what evidence would tell you whether that person can do the next job, because the organization has never separated the two ideas.
The essentials
- Performance is what someone delivers in the job they hold now. Potential is the likelihood they succeed in a materially different and larger job.
- They overlap, and they are not the same population. Treating them as one is the most expensive routine error in talent management.
- Performance data is abundant and backward looking. Potential has to be assessed deliberately, because nothing in your systems captures it.
- The cost of confusing them is paid twice: a weak manager, and a strong contributor lost.
What performance measures
Performance is a record. It says this person met or exceeded the commitments of their current role over a defined period, under conditions that existed at the time.
It is the more reliable of the two measurements and the more limited. It is reliable because it is observed rather than predicted. It is limited because it is bound to one job, one team, one market and often one manager who set the conditions for success.
Ask what a high performer's results actually depended on and the picture gets more interesting. A territory. A product cycle. An unusually capable deputy. A manager who removed obstacles. None of those travel with the person to a bigger role, and none of them are visible in the rating.
What potential means, and what it does not
Potential is a forecast about a different job. Not a bigger version of this job, a different one, with more ambiguity, more people, less direct control and a longer feedback loop.
The Center for Creative Leadership draws a line that most organizations lack a word for. It separates high potentials, who are broad and adaptable in their learning and skills, from high professionals, who have narrow but deep subject matter expertise. Both are high performers. Both deliver real value. CCL's warning is direct: do not confuse them, and the risk is highest in organizations that put a premium on technical skill.
That third category is the useful part. A principal engineer, a rainmaking partner, a plant manager who has run the same complex site brilliantly for a decade: these are not failed high potentials. They are a different asset, and most companies have no way of honoring them except by promoting them into management, which is how the asset gets destroyed.
The three things every credible model of potential asks about
Commercial models differ in their vocabulary and converge on roughly the same three questions.
Can they learn fast in unfamiliar conditions? Not intelligence, which they almost certainly have, but what happens when the familiar playbook stops working. Do they seek disconfirming information, or do they run the old play harder? This is the closest thing to a single predictor.
Do they actually want the next job? Aspiration is asked about far too rarely, and it is the cheapest thing to check. Plenty of excellent people want more scope and no more people, or more money and no more travel. A person who does not want the job will not do it well, and their reluctance is usually visible only after the appointment.
Can they get work done through other people? Every step up trades direct control for indirect influence. Someone who has never built anything through a team, or who has never persuaded a peer they do not control, has not yet been tested on the thing the next job is mostly made of.
Why your process pushes the two together
The confusion is not carelessness. It is built into how the work happens.
Performance data arrives automatically. It is in the system, it is quantified, it is defensible in a room. Potential requires an assessment somebody has to commission, a structured conversation somebody has to run, and a judgment call somebody has to defend. Under time pressure, the available data wins.
Incentives push the same way. A manager who names their strongest contributor as a high potential is signaling that they develop talent. Naming a solid but less visible performer with a genuinely broader profile invites the question of why. The path of least resistance is to put forward the person with the best numbers.
And there is a real cost to being right. Correctly identifying that your best performer is not a future general manager means telling them, or deciding not to, and both are uncomfortable. Many organizations avoid the conversation by never making the distinction in the first place.
How to tell them apart without buying anything
You can get most of the way with three moves and no vendor.
First, separate the two ratings on the form. If your talent review captures one score, you will get one concept. Two axes force two conversations, which is the mechanism behind a well-run 9 box talent review.
Second, ask for evidence outside the current role. What has this person done that their job did not require: a cross-functional problem they took on, a team they built in a domain they did not know, a position they changed after hearing an argument. Absence of evidence is not proof of absence, but it tells you where to look next.
Third, ask the person directly what they want, in a conversation that is explicitly not a promotion discussion. The answers are frequently surprising and always cheaper to hear now.
Where an instrument earns its cost is at the top of the house, when the appointment is expensive and you need a view on how the person will behave under strain rather than on a good day. That is a different question from potential, and we set out the options in Hogan vs DISC.
What to do with a high performer who is not a high potential
Start by not treating it as bad news, because it is not. A specialist who is excellent at a hard job is worth more to most companies than an average general manager.
Build somewhere for them to go that is not management. Senior individual contributor levels with real compensation bands, titles the organization respects, and scope that grows through complexity rather than headcount. If your top individual contributor band pays less than your first-line manager band, you have already told everyone what you value.
Use them where their depth compounds: standards, hard client relationships, technical arbitration, mentoring. And be honest in the conversation. A person told plainly that they are valued as the best in the company at a difficult thing, with a path and a number attached, generally takes that better than a vague promise of leadership someday.
The internal frictions that keep this muddled
The first friction is that the manager making the call is also the person who benefits from keeping the answer vague. A manager with a strong performer has every reason to label them high potential internally, because it protects budget and reflects well, and no reason to state plainly that the person is a specialist. The label costs nothing to apply and the consequence lands two years later on somebody else.
The second is that most organizations have exactly one currency of recognition, which is promotion into management. When advancement, pay and status all move along a single track, telling someone they are not on it reads as a demotion however carefully it is phrased. That is a structural problem in the job architecture, not a communication problem, and no amount of messaging fixes it.
The third is that nobody carries the cost of a bad potential call. The manager who nominated has moved on. The human resources partner who ran the process has a new portfolio. The person who inherits an underperforming team two years later has no way to trace the decision, and so the same mechanism runs again. Assessment quality does not improve without someone reviewing, at least annually, how the last set of calls turned out.
How you will know in ninety days
Ninety days is too short to judge a leader and long enough to test whether the assessment was sound. Look at three things.
Watch what they do with what they do not know. A correctly identified high potential asks more questions in the first month than they did in their last year, and changes something in response. Someone promoted for past performance tends to do the old job in the new title, visibly and with impressive output.
Watch who gets the credit. Within a quarter, the person's team should have started appearing in their updates by name. If every win is still first person singular, the shift from performing to enabling has not started.
Watch the quiet indicators. Has anyone on their team raised a problem with them and been heard? Have they made a decision that cost them something in the short term? Neither shows up in a dashboard, and both predict the next two years better than the numbers do.
If none of those has moved by day ninety, the call was probably wrong, and it is a great deal cheaper to add support now than to run the experiment for another year. Working out which of your people are which, on evidence rather than on last year's numbers, is what our leadership assessment work is for.
Frequently Asked Questions (FAQs)
Can a high performer also be a high potential?
Yes, and many are. Sustained strong performance is generally a precondition for being considered, because it demonstrates reliability and earns the person credibility. The point is that performance alone is not sufficient evidence, and treating it as sufficient is where the error occurs.
Can someone be a high potential without being a high performer?
Occasionally, and it deserves scrutiny rather than dismissal. A capable person in the wrong role, under a poor manager, or newly arrived in an unfamiliar domain can show real potential while their current numbers lag. Look for whether the underperformance is situational and time bound before you write them off.
What percentage of employees should be labeled high potential?
Practice varies widely and there is no correct figure. What matters more is that the label carries a consequence: real development investment, real exposure, and a real review date. A list that includes a fifth of the company and changes nothing is a list, not a program.
Should you tell people they are on the high potential list?
There are defensible positions either way, and the worst option is being unclear. Telling people raises expectations you then have to meet, which is a genuine cost. Keeping it secret rarely works, because the development investment is visible, and people draw their own conclusions from who gets the assignment.