A high-performing employee is often the most visible candidate for promotion.
The logic seems reasonable.
A scientist consistently produces excellent work. An architect solves difficult design problems. An engineer becomes the person everyone approaches when something breaks. A salesperson repeatedly exceeds targets.
When a managerial position opens, promoting that person can feel like the obvious next step.
But there is a problem with that assumption:
being exceptional at doing the work is not the same as being exceptional at enabling other people to do the work.
Management is not simply a more senior version of individual contribution.
It is a different kind of responsibility.
And as AI changes how work is distributed between people and machines, that difference may become even more important.
A promotion can actually be a change of profession
A great technocrat may have spent years developing deep expertise.
Their success may come from being able to analyse complex information, solve difficult technical problems, work independently and produce consistently high-quality outcomes.
Then they become a manager.
Suddenly, much of their success depends on very different things.
They need to delegate.
They need to set priorities when several things matter at once.
They need to manage conflict.
They need to communicate expectations.
They need to develop people who do not think exactly as they do.
They need to make decisions with incomplete information.
They need to balance resources, timelines and competing interests.
And sometimes they need to resist the temptation to simply take the work back and do it themselves.
The technical expertise has not stopped being valuable.
But it is no longer enough.
Research gives this problem a useful historical foundation.
Alan Benson, Danielle Li and Kelly Shue studied sales workers across 214 firms and found evidence consistent with the “Peter Principle.” Firms strongly rewarded current sales performance when promoting people into management, even though other observable characteristics were more predictive of managerial performance.
The researchers' broader finding was straightforward: the strongest individual performer was not necessarily the person with the strongest managerial potential.
The capabilities that create individual success can be different from those that create team success
Consider a brilliant engineer.
When a difficult technical problem appears, the engineer may be able to solve it faster than anybody else.
That makes them extremely valuable.
But once they lead a team, the question changes.
Can they help another engineer become better at solving the problem?
Can they give useful feedback without taking control of the work?
Can they recognise when somebody needs guidance and when somebody needs autonomy?
Can they handle disagreement constructively?
Can they make the team stronger rather than simply remain the strongest person on the team?
These are different capabilities.
The same distinction applies across professions.
A great scientist is not automatically a great research leader.
A great architect is not automatically a great project or people manager.
A great salesperson is not automatically a great sales manager.
And an exceptional technical specialist may be far more valuable remaining an exceptional technical specialist.
This does not diminish technical excellence.
It simply recognises that managerial performance should be evaluated against managerial requirements.
Where organisations want structured information about capabilities such as decision-making, communication, conflict management, delegation and people management, Psychometrica's Managerial Skills Profiler is one example of an assessment designed around managerial competencies. Managerial Skills Profiler
It should add to evidence from performance, experience, interviews and organisational context—not replace them.
Management is becoming more about leadership than supervision
Management has traditionally included planning, coordination, resource allocation and monitoring.
Those responsibilities remain.
But the idea of a manager as somebody whose main role is simply to supervise the work of others is becoming increasingly outdated.
Knowledge workers often know more about their own specialist areas than their managers do.
Teams may work across locations and disciplines.
Projects move quickly.
People need enough autonomy to make decisions without waiting for continuous instruction.
That means managers increasingly succeed through clarity, trust, judgment, communication and influence.
The World Economic Forum's Future of Jobs Report 2025 reflects this wider shift in employer priorities.
In its global survey, analytical thinking remained the most commonly identified core skill. But resilience, flexibility and agility ranked second, while leadership and social influence ranked third, identified as a core skill by 61% of surveyed employers.
Motivation and self-awareness, empathy and active listening, talent management and lifelong learning also appeared among the skills employers considered increasingly important.
Technical capability clearly still matters.
But organisations also value the distinctly human capabilities required to make groups of people work well together.
AI is changing the manager's job again
AI adds another layer to this transformation.
Managers are increasingly unlikely to manage only people.
They may also manage workflows in which humans work alongside AI systems and autonomous or semi-autonomous agents.
That changes the managerial question from:
“Who on my team should do this?”
to something more complex:
“What should a person do, what can AI do, where should an agent execute the work, and where does human judgment still need to remain?”
Microsoft's 2026 Work Trend Index describes organisations moving further toward work involving AI agents.
Its research argues that as agents take on more execution, people increasingly direct work, evaluate results and remain responsible for outcomes.
The report also says managers have an important role in operationalising an organisation's AI strategy. In a separate Microsoft-led study of 1,800 workers, employees whose managers actively modelled AI use reported greater AI value, critical thinking around AI and trust in agentic systems.
Microsoft is careful to describe these results as associations rather than proof that managerial behaviour alone caused those outcomes.
Still, the direction is interesting.
The manager increasingly becomes an orchestrator of capability.
Managing an AI agent and managing a person are not the same thing
This distinction deserves attention.
An AI system does not need career development.
It does not need recognition.
It does not need psychological safety.
It does not feel that its contribution has been ignored.
It does not need a manager to recognise that it is becoming disengaged.
A human colleague does.
An AI agent instead needs things such as:
clear objectives,
appropriate context,
access and permissions,
defined quality standards,
monitoring,
evaluation,
and escalation boundaries.
As agentic work expands, organisations also need people to review outputs and decide when systems should be changed.
Microsoft's 2026 research explicitly raises questions such as who reviews agent performance, who can update agent workflows and how accountability is maintained when automated work scales.
The future manager therefore may need to understand two fundamentally different kinds of resources:
human capability and machine capability.
Knowing how to use one does not automatically mean knowing how to lead the other.
AI may make human qualities more visible, not less important
It is tempting to assume that as AI becomes better at analysis, drafting, reporting, coding and other cognitive tasks, human capabilities will matter less.
A different possibility is emerging.
When machines perform more execution, people spend more time deciding:
What should be done?
What outcome actually matters?
Is this output correct?
Is it appropriate?
What does the context change?
Who should be involved?
When should the system be challenged?
Who remains accountable?
These are questions of judgment.
And management contains another category of questions that technology does not simply make disappear:
How do I tell someone that their performance is not meeting expectations?
How do I help a capable employee regain confidence?
How do I resolve conflict between two strong performers?
How do I recognise that someone is overloaded before they burn out?
How do I create an environment where a junior employee can challenge a senior one?
How do I make people feel that their work has meaning when technology is changing what their roles look like?
These are deeply human parts of management.
AI can assist managers with information.
It may help prepare performance summaries, analyse project data, generate options or identify patterns.
But the existence of better tools does not eliminate the responsibility to exercise judgment, build trust and lead people.
In that sense, AI may make the human part of management easier to see.
One of the hardest managerial skills is knowing when not to do the work yourself
High performers often become successful precisely because they are good at taking ownership.
Something difficult appears.
They solve it.
That behaviour may have been rewarded throughout their career.
Management can require the opposite instinct.
Imagine a manager watching a junior employee struggle with a problem the manager could personally solve in twenty minutes.
Doing it themselves may produce the fastest immediate result.
But repeatedly doing so can create a dependent team.
The short-term optimisation damages long-term capability.
A good manager therefore needs to ask:
Should I solve this?
Should I coach someone through it?
Should I delegate it completely?
Should the team collaborate?
Could AI perform part of it?
Does the output require human review?
Where does my own judgment add the most value?
This ability to allocate work intelligently across self, team and technology may become an increasingly important managerial capability.
Organisations can also create the wrong managers through their career structures
Sometimes the problem does not begin with the employee.
It begins with the organisation.
Many companies still imply that career progress means moving from:
technical contributor
to senior contributor
to manager.
Eventually, someone who wants more status, compensation or influence can feel that becoming a manager is the only way forward.
That creates an unfortunate possibility.
The organisation takes an exceptional specialist out of the work they do exceptionally well and places them into a role for which they may have less interest or aptitude.
The result can be:
the loss of an excellent specialist and the creation of an ordinary manager.
Strong organisations therefore need credible expert career paths alongside managerial ones.
Becoming a manager should not be the only way to demonstrate advancement.
And promotion should not simply answer:
“Who has earned it?”
It should also answer:
“Who is suited to the work this new role actually requires?”
Managerial potential deserves to be evaluated separately
Once organisations accept that individual and managerial performance are different, the selection process also needs to change.
Past performance still matters.
Technical expertise still matters.
Experience still matters.
But managerial roles may require additional evidence around areas such as:
decision-making,
communication,
delegation,
people management,
conflict management,
change management,
problem-solving,
judgment,
adaptability,
and the ability to influence and develop others.
For organisations evaluating employees for broader leadership responsibilities, Psychometrica's Manager BluePrint combines managerial competency assessment with workplace personality profiling as one source of additional information. Manager BluePrint
Again, an assessment should not decide who becomes a manager.
The better use is to give decision-makers additional structured information that can be considered alongside actual performance, interviews, experience, organisational needs and professional judgment.
The best performer may still become the best manager
None of this means organisations should avoid promoting high performers.
Many excellent individual contributors become excellent managers.
Their technical credibility can help them enormously.
Their understanding of the work can make them better coaches and decision-makers.
Their history of strong performance may demonstrate discipline, intelligence, judgment and commitment that remain valuable in leadership.
The point is simply that these conclusions should not be assumed.
Strong individual performance tells us that someone has succeeded at their current work.
Managerial potential asks a different question:
Can this person create the conditions in which other people can succeed?
The definition of a good manager is expanding
The manager of the future may still need to manage budgets, timelines, projects and resources.
But increasingly, they may also need to:
lead people through constant change,
build psychologically safe teams,
develop human talent,
evaluate AI-generated work,
coordinate people and agents,
decide where automation is appropriate,
maintain accountability when machines execute tasks,
and preserve distinctly human qualities inside increasingly automated organisations.
That makes technical expertise valuable—but insufficient.
It also makes management harder to define through seniority or individual achievement alone.
The strongest individual contributor may be the person who produces the best work.
The strongest manager may be the person who enables people, systems and technology to produce better work together.
That is a different kind of capability.
And in the age of AI, organisations may need to become much better at recognising it.
Frequently Asked Questions
Why don't top performers always become good managers?
Individual contributors and managers perform different kinds of work. Strong individual performance may depend heavily on technical expertise and personal execution, while management can require delegation, communication, decision-making, conflict management, coaching and responsibility for other people's performance.
Is technical expertise still important for managers?
Yes. Technical or domain expertise can help managers understand the work, make informed decisions and earn credibility. The issue is not that technical expertise is unimportant, but that it may not by itself demonstrate managerial capability.
What is managerial potential?
Managerial potential broadly refers to the capabilities and behavioural characteristics relevant to performing a managerial role successfully. The specific requirements vary by organisation and role but can include decision-making, communication, people management, delegation, adaptability, conflict management and judgment.
How is AI changing the role of managers?
Managers increasingly work in environments where AI tools and agents participate in workflows alongside people. This can require managers to decide which work should be performed by humans or AI, establish quality standards, evaluate AI-assisted outputs and maintain accountability while continuing to lead and develop human employees.
Will AI make human management skills less important?
Not necessarily. As AI handles more execution and information processing, skills such as judgment, communication, trust-building, leadership, coaching and ethical responsibility may become more visible parts of a manager's contribution. Current employer surveys continue to rank leadership and interpersonal capabilities among important workforce skills.
Should the best employee be promoted to manager?
Strong performance should be considered, but it should not automatically determine promotion. The better question is whether the employee's capabilities, interests and behaviours align with the demands of the managerial role.
Can psychometric assessments identify good managers?
Relevant assessments can provide structured information about specific managerial competencies or behavioural tendencies, but they should not be used as the sole basis for promotion or hiring decisions. Assessment results are better interpreted alongside performance history, structured interviews, experience, organisational context and professional judgment.
References
- Benson, A., Li, D., & Shue, K. Promotions and the Peter Principle. The Quarterly Journal of Economics, 2019. The study examined sales workers across 214 firms and found evidence that organisations can over-weight individual performance when making managerial promotion decisions.
- World Economic Forum. The Future of Jobs Report 2025 — Skills Outlook. Employers identified analytical thinking, resilience, leadership and social influence, creative thinking and self-awareness among the most important workforce skills.
- Microsoft. 2026 Work Trend Index: Agents, Human Agency, and the Opportunity for Every Organization. The report examines the growing use of AI agents and the role of managers, culture and organisational practices in enabling AI-supported work.
- Microsoft. 2026 Work Trend Index — State of Agents and Evaluation Infrastructure. The report discusses human oversight, agent performance evaluation, workflow ownership and accountability as agentic systems become more widely used.
