Every CHRO has seen the same graph.
You stand up a high-potential program. You identify 20 people. You invest heavily in them for 12 months. Two years later, eight of them have left. Of the ones who stayed, the promotion and progression data looks only marginally better than their non-HiPo peers.
Something went wrong. Something almost always goes wrong in HiPo programs.
Here is what we see.
Three failure modes
Failure 1: Prestige without trajectory. The program is a badge. Participants learn they are "on the list", but nothing in their day-to-day actually changes. No new work. No new stretch. No new mentor. The badge becomes a liability, it is a signal to other employers that they are ready.
Failure 2: Selection is opaque. Selection criteria are vague, managers nominate favourites, and the cohort skews toward people who are already visible. You miss quieter high-potential people, particularly women and people from under-represented groups. Trust in the program erodes.
Failure 3: Development is generic. You run a leadership curriculum that would suit anyone. The content is fine. But a person ready to run a $50M P&L and a person ready to run their first team do not need the same program.
Fix all three and retention outcomes change dramatically.
The selection move
The best HiPo selection processes share four features.
1. A published definition of potential. Write down, in public, what you mean by high potential. Ours typically has three axes:
- Capability ceiling: how much bigger could this role get in their hands over the next 24 months?
- Learning velocity: how fast do they absorb and apply new context?
- Judgement under ambiguity: do they make good calls when the data is thin?
Other definitions work. Opacity does not.
2. Manager nomination plus a second view. Nominations alone skew. Require a second perspective, a skip-level, a peer, or a structured behavioural assessment. You are looking for calibration, not two people to say the same thing.
3. Blind consideration of populations that get missed. Look deliberately at people the system tends to overlook. Part-timers. People returning from parental leave. People in less visible functions. Quiet high performers.
4. A published decline process. People who are not selected should know why, what the criteria were, and what would get them there next cycle. Without this, HiPo programs silently demotivate 80% of your workforce.
The development move
This is where most HiPo programs over-invest in content and under-invest in the thing that actually develops people: real work with real stakes.
Three design choices that change the trajectory:
1. A stretch assignment, not a rotation tour. A rotation tour exposes someone to multiple areas of the business. A stretch assignment forces them to own an outcome that is genuinely hard. One is sightseeing. The other is development.
Good stretch assignments:
- Own a P&L for 6 months
- Lead a cross-functional change the business actually needs
- Stand up a new capability (team, function, product line) from zero
Pair the assignment with a senior sponsor who meets with the HiPo monthly and holds them accountable to the outcome, not the process.
2. A single sponsor, not a crowd of mentors. A sponsor is different from a mentor. A mentor gives advice. A sponsor advocates for you in rooms you are not in, creates new opportunities, and puts their own reputation on the line.
Every person in the HiPo program should have one named sponsor at the executive level. That relationship is the single highest-leverage development lever you have.
3. Cohort learning on the work, not on abstract content. Bring the cohort together four to six times a year to work on their real problems together. Structured peer consulting, not a leadership curriculum.
In a typical session: each person presents one live problem they are facing, peers ask clarifying questions for 10 minutes, and offer three specific moves in the next 30 minutes. You leave with an action list, not a notebook.
This is both the best skill-building you can do and the stickiest retention mechanism: people stay for peers they respect and learn from.
The retention move
The honest truth is that a well-run HiPo program slightly accelerates attrition unless you actively manage the retention side. Visibility cuts both ways.
Three things help.
1. Talk about the career path every six months. Sit down with the HiPo, their sponsor, and HR. Ask: what are the 2 or 3 most likely next roles in 18 to 36 months? Are any of those actually available? If they are not, you have to either create them or be honest that this person will hit a ceiling.
If you cannot create the path, you are effectively training people for your competitors. That is sometimes the right answer. Just know that is what you are doing.
2. Pay at the top of the band, not the middle. People on HiPo programs are the most easily recruited people in your organisation. Do not pay them at the 50th percentile. Pay them at the 75th or higher, and make sure they know it.
3. Give them a say in the next move. The people who stay are the people who believe they have agency in where they go next. Build a real conversation about options into the program rhythm, not just a survey at the end.
What good looks like at 12 months
If your HiPo program is working:
- Selection criteria are published and used
- Every participant has a named stretch assignment with a clear business outcome
- Every participant has a single executive sponsor with a monthly cadence
- Cohort sessions are on the work, not on generic leadership theory
- Retention in the HiPo population is materially higher than comparable non-HiPo
- The business has backfilled 2 to 3 significant roles from the program
If the only output of your program is that 20 people attended a residential, you are signalling prestige without changing trajectories. Redesign.
A starting point
Welcome to better.