She was the best individual contributor on the team. That is why she was promoted. The logic was so obvious nobody examined it: she shipped the most, she understood the work most deeply, she was the person others already went to when something broke. Making her a manager was less a decision than a reflex. The offer letter changed her title, her band, and her compensation. It did not change a single thing she knew how to do.
Eight months later, her best engineer resigns. In the exit conversation he says, carefully, that he had stopped growing. That he had not had a real conversation about his development in over a year. That his manager was responsive, available, technically excellent, and completely absent on the one thing he needed. None of this is in her performance review. Her team's delivery metrics are strong. She is, by every number the organisation tracks, succeeding.
She is failing at the part of the job nobody ever taught her, that nobody measures, and that the system she works inside actively punishes her for prioritising.
The Skill Nobody Handed Over.
Start with the obvious thing the organisation pretends is not obvious. Developing people is a discrete, learnable, difficult skill. It is not a personality trait. It is not seniority expressed in a softer register. It is the specific capability to look at another person's work, see the gap between where they are and where they could be, and design the experiences, feedback, and stretch that close it without breaking them. Almost nobody is born able to do this. Almost nobody is taught it before they are made responsible for it.
What the new manager was taught, relentlessly and well, was how to do the work herself. Her entire career rewarded a single instinct: when something is hard, lean in and solve it personally. That instinct made her promotable. It is also precisely the instinct that destroys a developing engineer, because the fastest way to deny someone growth is to keep taking the interesting problem off their desk and finishing it yourself. The behaviour that earned her the role is the behaviour the role requires her to unlearn. The organisation handed her a job that demands the opposite of everything it spent a decade reinforcing, and then offered her a half-day workshop on giving feedback.
The workshop, predictably, did nothing. Not because it was badly designed. Because a skill this load-bearing cannot be installed in an afternoon, and because the moment she returned to her desk the system around her resumed paying her, promoting her, and praising her for output, not for development. She is a rational actor. She optimises for what is measured. Development is not measured, so development is what gives way.
You cannot delegate the development of people to managers who were selected for, trained in, and rewarded for the opposite skill, and then act surprised when development does not happen.
The Three Things the System Never Gave Her.
Walk through what would actually have to be true for this manager to develop her people, and the scale of the organisation's quiet abdication comes into focus.
First, she would need the capability. Real capability, built deliberately over months, not a slide deck on coaching models. She would need to have watched someone good do it, tried it under observation, gotten it wrong, and been corrected. The organisation provided none of this. It assumed the capability would materialise from good intentions and a sense of responsibility, which is roughly the same as assuming someone will be able to perform surgery because they care about the patient.
Second, she would need the time. Developing people is slow, unglamorous, and invisible in the quarter it happens. It is the one-to-one that runs long because the real issue finally surfaced. It is the deliberate decision to let a junior engineer struggle with a problem for three days instead of solving it in twenty minutes. Every one of those choices costs short-term throughput, and her throughput is the only thing her own manager asks about. The system gave her a developmental mandate and a delivery scorecard, and the scorecard wins every single time, because the scorecard is the thing that determines her next promotion.
Third, she would need permission. Permission to count the hours she spends growing someone as real work rather than a tax on the real work. Permission to be told, by the level above her, that a quarter in which she shipped slightly less but developed two people into genuine seniors was a good quarter. That permission was never granted, because the level above her was promoted under the identical logic and is running the identical playbook. The failure is not personal. It is inherited. It compounds downward through every layer of the organisation, each manager modelling for the next that development is the thing you talk about in reviews and abandon in practice.
The Reframe Nobody Wants on the Slide.
Here is the part that should make the room uncomfortable. The problem is not that this manager is bad at developing people. The problem is that the organisation has defined the manager's job in a way that makes developing people irrational, and then located the failure in the individual.
Every conversation about this defaults to the person. She needs more coaching. She needs a development plan. She needs to be more intentional with her one-to-ones. Send her on the leadership programme. All of it treats a structural failure as a personal deficiency, which is convenient, because fixing a person is cheaper and less threatening than fixing a system. It is also why the same conversation will happen again next year about a different newly promoted manager, with the same diagnosis and the same useless remedy, in perpetuity. An organisation that keeps producing managers who cannot develop people does not have a manager problem. It has a manager-production problem, and the production line is working exactly as designed.
The competency framework will tell you development is a core expectation of every people manager. The promotion criteria will quietly tell you the truth: that nobody has ever been promoted for it, and nobody has ever been denied promotion for failing at it. People do not do what the framework says. They do what the promotion rewards. The gap between those two documents is the gap your best engineers are resigning into.
What Rebuilding Looks Like.
A manager who develops people is not a better-intentioned manager. It is a manager standing inside a system that was built to make development the rational choice rather than the heroic one.
That system begins before the promotion, not after it. The decision to move someone from contributor to manager is treated as a genuine transition into a different profession, with the capability built deliberately ahead of the title, not bolted on as remediation once the damage is visible. The question stops being who is the best engineer and becomes who can grow other engineers, and the organisation accepts that these are frequently not the same person, and that a brilliant contributor kept brilliant and uncrowned is often worth more than the same person made into a mediocre and resentful manager.
It continues with measurement that has teeth. If development is not on the scorecard that determines pay and promotion, it does not exist, no matter how prominently it appears in the values deck. The organisations that actually develop people make the growth of a manager's team a visible, weighted, consequential input into that manager's own advancement. They track whether people under a given leader get more capable over time, and they treat a leader whose best people stagnate or leave as a leader with a problem, not as a leader with strong delivery numbers.
And it requires the hardest admission of all, which is that this is an architecture question and not a training question. You do not fix it by sending managers on a course. You fix it by changing what the system selects for, what it measures, what it rewards, and what it tolerates, so that a manager developing her people is swimming with the current instead of against it. This is the work SSUNDAR is built for. Our Performance Systems Architecture engagements rarely begin with a request to teach managers how to coach. They begin with a leader who has run that course three times, watched it change nothing, and is finally ready to look not at the managers who keep failing but at the system that keeps manufacturing the failure. The intention was never missing. The architecture to hold it was.
She was never bad at developing people. She was promoted by a system that taught her the opposite, paid her for the opposite, and then wrote her up for the gap.