One of the quiet disappointments of seniority is discovering how little of it applies to the people around you.
You build judgment. You develop a point of view on what good looks like. You get clearer, over time, about the kind of people who make work better and the kind who make it harder. And then you show up on Monday and the cast is whoever it is — assigned by org design, by history, by a hiring decision made before you arrived, by a stakeholder map you had no hand in drawing.
The question is not whether you would have chosen them. You didn't get that option. The question is what you do with who you actually have.
Inside your team, you have more choice than you think — and less than you want.
You can hire. You can develop. You can set standards, hold people to them, and create the conditions where the right people stay and the wrong ones leave. This is real leverage, and most leaders underuse it — either by hiring for capability without hiring for character, or by tolerating misalignment long after the evidence is clear because the conversation is uncomfortable.
The team you keep is the team you are choosing. Every quarter you do not address the person who is technically competent but corrosive to everything around them, you are choosing them again. Inaction is a decision. It just feels like not having made one.
But even inside the team, the choice is constrained. You inherit people. Org structures shift. Someone is moved into your scope without your input. The question then is not whether they were your choice — it is what standard they are being held to now that they are.
Outside the team, you have almost no choice at all.
The stakeholder who is difficult to read. The peer who takes credit and distributes blame. The senior leader whose support is visible in public and absent when it actually matters. You did not hire them. You cannot performance-manage them. You have to work with them anyway — and what you produce together will be judged as if the relationship were functional, regardless of what it actually is.
This is the part of leadership writing that most advice skips, because the honest answer is unsatisfying. You cannot choose your stakeholders. You cannot change people who do not see a reason to change. You can only control the terms on which you engage — and that control, exercised with precision, is more than it sounds.
Then there is the harder category.
The person who presents well. Who says the right things in the right rooms. Who is, by all visible measures, a constructive presence — until something is at stake, and then the behaviour shifts in ways that are hard to name and harder to prove. The credit quietly taken. The commitment made and quietly not kept. The support offered in the meeting and withdrawn in the follow-up. The version of events that reached the senior leader that did not quite match what happened.
These are not dramatic villains. They are harder than that. They are people who have learned to operate at the boundary of what is legible — where the behaviour is just ambiguous enough to survive scrutiny, and the person raising it looks like the difficult one for raising it.
The first thing to know is that you are probably not the only one who sees it. These patterns rarely target only one person. Somewhere in the org, someone else has noticed the same gap between the public version and the operational one. You may never coordinate explicitly, but you are not alone in the reading.
The second thing is to document without drama. Not a case file. A habit. Written follow-ups after every consequential interaction. Decisions confirmed in email. Commitments named with owners and dates. This is not paranoia — it is the operating standard recommended for any high-stakes cross-functional environment. Applied consistently, it is not a signal that you distrust someone. It removes the ambiguity that behaviour like this depends on.
The third thing is to reduce the surface area of exposure. Not avoidance — calibration. The person who is unreliable gets fewer opportunities to be unreliable on something that matters. You route around the dependency where you can. Where you cannot, you double the verification and halve the assumption.
The fourth is to name the pattern to the right person — once, cleanly, as a business problem. Not "I don't trust this person." Something more specific: "There has been a recurring gap between what is committed and what is delivered, and it is affecting the timeline. I want to flag it before it becomes a larger issue." One conversation. The right person. Evidence, not frustration.
Staying sane in this context is less about managing them and more about managing what you allow it to cost you.
The leader who spends significant energy trying to decode, outmanoeuvre, or change someone operating in bad faith has made a strategic error. Not a moral one — the frustration is completely understandable. But attention is finite, and attention spent on a person who will not change is attention taken from the work, the team, and the problems that are actually solvable.
The sanity is in the boundaries, not the battles. Know what you will and will not allow. Make the record clean. Protect the team from the worst of it where you can. And then — and this is the part that takes practice — return your attention to what you can actually build.
You did not choose the cast. You rarely do, at any level of seniority. What you choose is the standard you hold yourself to inside whatever cast you were given. That is the only variable entirely in your control.
You cannot choose the people you inherit. You can choose the terms on which you engage, the standards you hold, and how much of your attention the unchangeable ones get to consume.