One of the worst things I have ever done to a good employee was give him a promotion he had earned. He was the strongest technician we had, he had been with us longer than almost anyone, and when a management position opened up it was obvious to everybody, including me, that it belonged to him. I gave it to him. Within a year he was miserable, the team around him was worse off than it had been before, and I had lost the best technical person in the building without gaining a manager to show for it. That one was mine, not his. I never asked whether he actually wanted to manage people, and I never looked for a single piece of evidence that he could.
The setup is nearly always the same, and I would guess a number of you are sitting in some version of it right now. A management seat is coming open and you already know which of your senior people is going to want it. She has been with you longer than most of the team. She is exceptional at the work. She is a good employee who you trust, which in a small company counts for a great deal more than it does in a big one. The person leaving the role was an experienced manager running a team of real size, so the seat is not a small one. Here is the part worth sitting with: if she walked in as an outside candidate with no leadership experience anywhere on her resume, you would not shortlist her. Something about the fact that she is already yours makes applying that same standard feel like disloyalty.
The reason this goes wrong so often is that we are not promoting somebody, we are changing their profession. You have a person at the top of their game in one discipline and you are handing them a job that runs on an entirely different, frequently incompatible skill set. The thing that made her great, going deep on a hard problem until it yields, is close to the opposite of what the new job asks for, which is stepping back from the problem to look after the people working on it. On top of that she now has to manage a group of former peers, which is a difficult position for anyone, and impossible for some. The worst outcome is not that she fails. It is that you take the right person out of the right seat and put them in the wrong one, and lose them from both.
Anyone who has spent time in Traction, Gino Wickman's book on running a company on a single operating system, will recognize where I am going with this. It has a tool called the people analyzer, and at its core are three questions about anybody in any seat: do they get it, do they want it, and do they have the capacity to do it. The tool is well known and I am not going to improve on it here. What I want to add is the part that took me a long time to learn, which is that these are not questions you can put to the person directly. Ask her whether she gets it, wants it and can do it, and you will get an honest answer to a different question, which is whether she believes she can. You have to look at behaviour instead, and nearly all of the behaviour you need has already happened.
Does she get it. This is the one that trips people up, because it sounds like a question about intelligence and it is not. Getting it means understanding that the job is now to balance what the company needs, what the client needs and what the team needs, and that those three things pull against each other most days of the week. The place to look is what she does when she brings you a problem. Does she arrive expecting you to solve it, or does she come in with two or three ideas about what to do and a preference among them. When she describes the problem, who is in the description. Is it the client and what this costs them, the team and what it is doing to them, the company and what happens if this keeps recurring. Compare that with the version where what fills the description is her own workload, whether the thing is technically her job, and the most interesting technical route through it. None of those are bad instincts in a technician. They are simply not the instincts of the job you are about to offer her.
Does she want it. Most owners think they have already tested this, because she said yes. She said yes to a raise and a better title in the only direction your organization currently offers them, which is not the same thing. The cleanest way I know to separate the two is to put a real alternative on the table. Offer her two options: the leadership role, or a more senior technical track, with the same money and a title she would be equally happy to have either way. Given a genuine choice where the compensation is identical, she will pick the one she actually wants rather than the only one on offer. That conversation is worth having even when you are sure of the answer, and it is worth building the technical track whether or not she takes it, because the alternative is a company where the only way to earn more is to stop doing the thing you are good at.
Does she have the capacity to do it. This one has the most visible evidence attached to it, and it is the evidence you already have. How does she handle it when a junior person is stuck, does she coach them through it or take the keyboard. Can she deliver a hard message to a peer, or to a client who is in the wrong, without either flinching or turning it into a fight. When something is going badly on the team, do people go to her first, and if they do not, ask yourself why not. Does she take things on before anybody assigns them to her. You should be able to answer all of that from what you have watched over the past two years, and if you cannot, that is worth noticing on its own, because it usually means you have not been paying the kind of attention this decision requires.
If the answer in any of these three areas is no, the decision needs rethinking, and the three are not equal. Capacity is the one you can build. Coaching, difficult conversations, running a meeting that goes somewhere, holding somebody to a commitment, all of that is learnable, and part of your job if you promote her is to teach it or to buy the teaching. The first two do not work that way. Wanting the job and understanding what the job is are closer to instincts than trained behaviours, and you cannot coach somebody into wanting a life they did not want on Monday. Getting this wrong is expensive in a way that shows up nowhere in your numbers for a year or more, because a struggling manager does not fail loudly. The team just quietly gets worse, and the people who leave over it give you a different reason on their way out.
There is a version of this decision that goes well, and it usually happens in companies where the owner started watching for these things long before a seat came open. The person who was going to be a good manager was already coaching people nobody asked her to coach, already framing problems in terms of what they cost the client, already having the awkward conversation instead of routing it to you. Those are not things she started doing when the job was offered. They are things she had been doing for a year, and all you are doing at that point is confirming what you have already seen. That kind of visibility into your own team is a large part of what building a real leadership team is actually made of, and it depends on having handed over enough genuine authority that people had the room to show you who they are.
If you are looking at this decision right now and you are not sure, that uncertainty is information. Take the time to look at the evidence you already have before the seat gets filled, and if you want an outside read on it, that is a conversation worth having before the offer rather than a year after it.