Every company has a high-potential list. Some are written down; most live in the heads of three executives and change depending on who presented well last quarter. The written-down kind is better, but only if the criteria underneath it measure what the label claims: capacity for bigger roles, not comfort in the current one.
This is a practical guide to building a HiPo identification process that managers trust, candidates survive, and the succession plan can actually use.
Performance and potential are different questions
Start by separating the two questions your process must answer, because conflating them is the root failure of most HiPo lists.
Performance asks: how well does this person deliver the job they have? It is backward-looking, comparatively easy to measure, and already handled by your review cycle.
Potential asks: how much bigger or more complex a job could this person grow into? It is forward-looking, harder to measure, and not derivable from performance alone.
The classic proof is the brilliant individual contributor promoted into management who flounders within two quarters. High performance in the old role, low potential for the new one, and the org lost a great specialist to create a struggling manager. The reverse case exists too and is more expensive: the solid-but-unspectacular performer with real capacity for scale, invisible to any process that filters on current ratings first.
Treat them as two axes, not one. That is the entire logic of the 9-box grid, and it is worth preserving even if you never draw the grid.
What to actually look for
Potential resists direct measurement, but it leaves fingerprints. The signals worth structuring your criteria around:
- Learning speed in unfamiliar territory. Not intelligence in the abstract: the observable rate at which someone gets competent when dropped into something new. The person who took over an unfamiliar domain and was fluent in a quarter is showing you the trait directly.
- Behavior when the plan breaks. Bigger roles are mostly ambiguity management. Watch who stays functional, keeps deciding, and keeps their team moving when the situation has no playbook. Watch who freezes or escalates everything upward.
- Pull from other leaders. When cross-functional peers keep requesting the same person for hard problems, that demand signal is hard to fake and hard to astroturf.
- Appetite, stated and revealed. Some excellent people do not want a bigger job, and their preference deserves respect rather than a spot on a list they'll decline. Ask. Then check whether behavior matches the answer: does this person volunteer for stretch, or negotiate scope down?
- Range of followership. Leading a team of people like yourself is one skill; getting results from people whose disciplines, incentives, and vocabularies differ from yours is the skill bigger roles run on.
Notice what is absent from the list: polish in executive meetings, tenure, and being liked by the CEO. All three correlate with getting named to HiPo lists. None of them predicts scale.
The biases that corrupt HiPo lists
HiPo identification is a bias magnet, and the biases are boring and predictable, which is good news: predictable failure modes can be engineered against.
- Visibility bias. The people executives see present well get named; equally capable people two floors from the action don't. Structural fix: every org unit must nominate from its full population, and lists get challenged for line-of-sight skew.
- Similarity bias. Leaders spot potential in people who remind them of themselves at that age. The phrase to ban from calibration sessions is "I can just see them in the role," which translates to "they resemble the last person who had it."
- Recency bias. One great quarter puts someone on the list; one rough project takes them off. Require evidence spanning at least a year.
- Halo from performance. The top performer gets waved onto the list without anyone asking the potential question at all. Force the two-axis conversation on every name.
Make the criteria explicit and scored
Whatever signals you choose, write them down as scored criteria, the same discipline you would apply to succession assessment generally. A workable format:
| Criterion | What evidence looks like | | --- | --- | | Learning agility | Took on an unfamiliar domain and reached competence fast, with examples | | Ambiguity tolerance | Kept a team deciding and moving through an unplanned situation | | Cross-functional pull | Named by peers outside their own line as the person they request | | Aspiration | States interest in larger scope, and volunteers for stretch in practice | | Followership range | Has gotten results from teams unlike themselves |
Score each on a small scale with a sentence of evidence, calibrate across raters, and date everything. The evidence sentence is the anti-bias mechanism: "high potential, trust me" cannot be calibrated, but "took over the pricing migration mid-crisis and shipped it" can be interrogated by the room. The payoff for the discipline is measurable: DDI's Global Leadership Forecast 2025 found promotions were 1.6 times more likely to succeed when objective assessment informed the selection.
Expect a defensible process to land 5 to 15 percent of the population on the list. Much larger than that and you are measuring performance twice; much smaller and the bar is probably confused with readiness for a specific named role, which is a different question.
Should you tell people they're on the list?
The uncomfortable question every process eventually hits. Both answers cost something.
Telling people risks entitlement, awkwardness if they later come off the list, and a two-tier culture if the list leaks (it leaks). Not telling people costs more, quietly: your best people don't know you see a future for them, and recruiters make sure they hear it from someone. DDI's 2025 research found high-potential talent was 3.7 times more likely to leave within a year when managers weren't providing regular growth and development opportunities.
The workable middle: communicate investment, not status. "We are putting you in the leadership program, giving you the integration workstream, and pairing you with an executive sponsor" tells someone everything they need to know about their trajectory without minting a caste label. Meanwhile, named-successor conversations for specific roles happen separately, under the succession plan, where they belong.
Whichever policy you pick, pick it deliberately and apply it evenly. The worst outcome is the accidental one where some managers tell and others don't.
From HiPo list to succession bench
A HiPo list that ends as a list is a filing exercise. Its purpose is to feed the succession pipeline, and the handoff works like this:
- HiPo identification finds capacity for scale across the whole population.
- Succession planning matches that capacity against specific critical roles, scores candidates against role profiles, and assigns readiness windows.
- Development assignments close the specific gaps the role scoring exposed, and completed assignments generate the evidence for the next assessment pass.
The two processes check each other. A HiPo who scores poorly against every actual role profile is a flag on your criteria. A critical role whose entire bench came from one manager's team is a flag on your nomination funnel. When both processes run honestly, the succession bench stops being a list of usual suspects and starts being the org's real capacity, made visible.