Every organization wants to identify its future leaders as early as possible. Most do it through talent reviews, where leaders meet to discuss employees’ potential. But deciding who has potential also decides who gets development opportunities. So the key question is this: Do these reviews find the right people, or do they overlook capable ones?
Why is potential difficult to see?
Performance describes what someone has achieved in a particular role. Potential predicts how someone might perform in roles they haven’t held yet. Those roles often involve situations that haven’t occurred. Forecasting potential is harder than measuring past performance, and we often confuse the two.
Research by CEB, now part of Gartner, found that confusing high performers with high potentials is costly. Only about 15 percent of strong performers had what more senior roles demand. The research also found that only about one-third of organizations used valid methods to identify high potentials. Nearly half had no systematic process at all.
Psychologists Rob Silzer and Allan Church propose that potential has three layers:
- Foundational dimensions: relatively stable traits such as cognitive ability and personality.
- Growth dimensions: learning agility, adaptability and motivation to develop.
- Career dimensions: the leadership, functional and technical capabilities a specific future role requires.
Silzer and Church framed the key question as “Potential for what?” Yet many talent reviews never answer it. When the review doesn’t define potential, managers and leaders fill the gap with their own biases. They tend to rate people highly when they resemble existing leaders.
How does bias affect potential ratings?
A 2019 study in Frontiers in Psychology shows how easily these judgments change. Abigail Player, Georgina Randsley de Moura and colleagues ran two hiring simulations with nearly 300 participants.
Participants evaluated candidates for a leadership role based on leadership performance or leadership potential. The study found:
- When the candidates were men, participants valued potential more highly than proven performance.
- When the candidates were women, the preference reversed. Participants favored those who showed a track record of leadership performance.
- The researchers called this the “overlooked potential effect.” It benefits men and disadvantages women seeking roles that prioritize potential.
- Men were chosen for what they might achieve, while women had to show what they had already achieved.
The study used a simulation, which more clearly demonstrates the direction of bias than real-world scenarios. However, the key lesson remains valid. Teams should compare performance ratings with actual results. Unfortunately, many organizations fail to assess potential ratings against later outcomes. This oversight creates a gap that permits bias to influence decisions, often favoring the most visible individuals over the most competent.
Who should be in the high-potential pool?
The first decision in any talent review is who gets considered at all. That’s often where organizations overlook people without realizing it. Entry criteria should be broad, objective and few. They define eligibility, and the assessments that follow do the sorting.
Reasonable entry criteria include:
- Role level: everyone in positions that feed the target roles, such as managers one or two levels below director.
- Time in role: about two to three years in the current role to show a track record. Avoid longer requirements, which can exclude fast learners.
- Solid performance: a rating of “meets expectations” or higher. Requiring top ratings blurs the line between performance and potential.
- Expressed interest: a stated wish to lead. Aspiration is an important part of potential.
- No active performance or conduct issues.
Everyone who meets the entry criteria should go through the same assessment process.
Which assessments measure potential?
Organizations don’t have to rely solely on a manager’s instincts. Several assessments measure potential directly, and many are backed by strong research; below are some of them:
- Learning agility assessments measure how quickly someone learns from experience and applies those lessons in unfamiliar situations. Korn Ferry, which has researched the concept for about 30 years, considers it a key indicator of leadership potential. Much of this evidence comes from vendors, so ask for independent validation data.
- Cognitive ability tests predict job performance across many types of work. A 2022 reanalysis by Paul Sackett and colleagues found their predictive validity lower than previously believed, at about .31. These tests can also produce score differences between demographic groups.
- Structured interviews ask every candidate the same job-related questions and score answers against set criteria. The same 2022 reanalysis ranked them as the strongest predictor of job performance.
- Assessment centers and leadership simulations put candidates through realistic exercises that trained assessors score. A classic meta-analysis by Gaugler and colleagues found stronger links to later ratings of management potential than to job performance. They work best with more exercises and with psychologists as assessors.
These tools can also make selection structured and formal. A 2003 doctoral study of retail store managers compared three methods. An assessment center and a multi-rater appraisal both predicted performance better than supervisor ratings. The two structured methods showed no adverse impact against minority candidates, while supervisor ratings did.
However, assessments alone aren’t a complete solution. If you assess only the people managers nominate, the assessment inherits the nominations’ biases. Offer assessments to the entire eligible pool and combine several methods rather than rely on one score. Validate the tools against employees’ later performance and check outcomes for adverse impact every cycle.
The same caution applies to artificial intelligence-powered talent platforms. These tools can surface “hidden” high potentials that managers overlook. They can also learn from past promotion decisions and repeat any bias. Before relying on one, ask the vendor three questions: How did they validate it? What data trained it? How did they test it for adverse impact? And keep a person accountable for every final decision.
Can we identify high potentials more fairly?
These recommendations explain how to identify high-potential individuals, build a strong talent pool, and enhance the effectiveness of talent reviews.
- Ask “Potential for what?” before rating anyone. Define the specific future roles or capabilities you’re forecasting. Then identify the observable behaviors that signal readiness.
- Separate the potential and performance conversations. When discussed together, a strong year inflates perceived potential and a weak year deflates it. Rate them at different times, using different evidence.
- Replace single-manager judgment with structured, multi-source evidence. Combine the assessments above with input from several leaders who have directly observed the person’s work. Calibration panels should require evidence for every rating.
- Audit potential ratings the way you audit pay. Break down ratings by gender, ethnicity, age, location, working pattern, and function. Then track whether people’s later performance matched their potential ratings. Each year, KPI report representation at each pipeline stage, bench strength, and how accurately last cycle’s ratings predicted performance. Also report retention for people on and off the list.
- Make labels temporary and reversible. Treat a HiPo designation as a hypothesis with an expiration date, reviewed every year. Create clear ways to join the list mid-cycle and to leave it without lasting consequences.
- Broaden access to challenging assignments. Rotate high-visibility roles, project leadership, and time with senior leaders across a wider pool than the formal list.
Looking ahead
Can we identify future leaders without restricting who can become one? Only to some extent. Any judgment about potential is a forecast, and forecasts can carry bias. The practical answer is to keep making predictions while staying humble about their accuracy.
Treat potential ratings as predictions to test rather than facts to accept. Clearly define what leadership potential looks like, have multiple people assess it and compare their predictions with actual outcomes. Keep the list flexible and open for updates after your initial draft.
A structured, formal talent identification process helps organizations identify future leaders early and allows capable employees to demonstrate their abilities and shape their future.













