Your Manager Is Also Guessing

She said "let me think about it and come back to you," and I spent four days constructing an elaborate theory about what that meant.

Political cover. A conversation happening above me. Some consideration I wasn't cleared for. I built the whole edifice out of one sentence, because that's what you do when you assume the person across the table is operating from a document you haven't been shown.

On day five she came back and said, more or less: I genuinely don't know, I've never had to decide something like this, what do you think we should do?

And the floor moved slightly.

Not because she was incompetent — she was one of the better ones, which is the entire point of this piece. The floor moved because I understood, eight years into a career, that there was no document. I had spent those eight years addressing my questions to an institution I imagined standing behind each manager, and the institution was four people improvising in different rooms, each assuming the others had been briefed.

The number that reframes the job¶

The Chartered Management Institute, polling with YouGov across more than 4,500 workers and managers, found that 82% of those who enter management positions have not had any proper training. CMI calls them accidental managers.

Sit with the size of that. Not a minority. Not a problem confined to small firms without an L&D budget. Four out of five.

Now put it beside what the role is actually asked to carry. Gallup, drawing on engagement data covering 27 million employees and more than 2.5 million work units over two decades, reported:

So: the single largest identifiable factor in whether a team's experience of work is tolerable, handed to people who were given no instruction in how to do it. That's the structure. Everything downstream — the vague answer, the meeting that resolves nothing, the feedback that arrives as weather — is a symptom of it.

They were promoted for a different job¶

Here's the mechanism, and it isn't anybody's villainy.

Management is awarded as a prize for individual contribution. You were an excellent engineer, nurse, account manager, so here is a team. The skills that earned the promotion — depth, focus, personal execution — are close to orthogonal to the ones now required, which are almost entirely about operating through other people. Nobody sat anyone down between those two states.

Gallup is blunt about the hit rate. Their research found "about one in 10 people possess high talent to manage", that "nearly one in five (18%) of those currently in management roles demonstrate a high level of talent for managing others", and that companies "miss the mark on high managerial talent in 82% of their hiring decisions."

I want to flag something about those numbers rather than let them do a rhetorical trick on you. CMI's untrained figure is 82%. Gallup's mis-hire figure is 82%. These are different studies, different countries, different methods, measuring different things, and the coincidence means nothing. It reads like corroboration and it isn't one. I'm noting it because if I don't, the pairing does persuasive work it hasn't earned.

What the two do jointly support is narrower and sturdy: the selection process and the training process are both weak, independently. Companies pick the wrong people and then don't teach them. Either alone would produce strain. Together they produce a profession staffed largely by improvisers.

Most of them know¶

This is the part that surprised me most in the data, and it's the part that should change how you talk to your manager.

From the same CMI research:

Roughly four in five managers are, when asked privately by a pollster, telling you they're underequipped. That is not a profession of frauds convinced of their own brilliance. It's a large population of people who know they're guessing and have no available script for saying so out loud — because the role's unwritten job description includes projecting certainty, and a manager who says I don't know to the wrong person in the wrong quarter can watch it become a performance note. The written job description, meanwhile, is five words on a poster that predict nothing about the place.

The consequences of the gap are measurable and they land on everyone. CMI found 31% of managers and 28% of workers have left a job because of a negative relationship with their manager. Among workers who rated their manager ineffective, half planned to leave within the year — against 21% of those who rated their manager effective. Two and a half times the attrition, tracking a variable the company declined to invest in.

And training moves it. Managers who'd had formal training reported confidence at 83%, against 71% for those who hadn't. Not a transformation. But it's the cheapest lever in the building and it is mostly sitting untouched.

The turn¶

There's a version of this argument that I want to disown before someone else makes it on my behalf.

This is not "be nice to your manager, they're doing their best." Some managers are cruel, and untrained cruelty is still cruelty. The 82% figure explains incompetence; it does not excuse contempt, and a person who humiliates their reports has not made a training error. Nothing here should be read as a request for patience with that.

The useful version is diagnostic, and it's sharper.

You have probably been addressing your requests to an authority that isn't there.

When you ask for a decision, you're modelling your manager as a node connected to a system that holds the answer — strategy, precedent, a doc. When 82% of them entered the job untrained, the far better model is: a reasonably smart person with slightly more context than you, no training, four other things on fire, and a strong incentive to look decisive.

Feed that model, and everything about how you operate changes. You stop asking what should we do here? — which requires them to generate an answer from nothing while performing confidence — and start saying here's what I think we should do, here's what I'd need, tell me what I'm missing. Same decision, radically different demand. The first asks them to be an institution. The second asks them to be an editor, which is a job humans are actually good at.

This is the same failure of imagination as expecting the org chart to describe who really decides things. We keep reading formal structures as though something authoritative stands behind them, and then experiencing the gap as a personal betrayal rather than as the ordinary condition. The forty-minute meeting exists for reasons that have little to do with information; the manager's hesitation exists for reasons that have little to do with you.

And if you run a small business, none of this is spectator sport. The moment you make your first hire you become an accidental manager yourself, with less training than any of the 82% and no HR function to absorb your mistakes — a transition the standard advice about starting a business mentions barely at all.

What to do with this on Monday¶

1. 1

   Arrive with a draft, not a question

   The single highest-leverage change available. Replace "what should we do about X?" with "I'm planning to do X, here's why, tell me what I'm missing." You've converted a generation task into an editing task, and you keep the pen — which means you keep the shape of the outcome.

2. 2

   Ask what they're actually measured on

   Most people never ask, and most managers will answer. You are trying to find the constraint they're optimising under, because half of what reads as arbitrary is a manager routing around a target you can't see. Once you know it, you can make your proposal serve it, which is how requests start getting approved.

3. 3

   Give them a way to not know

   "No rush if you need to check with someone" costs you nothing and removes the requirement to perform certainty. You'll get straighter answers, and faster ones, because you've made the honest response cheap.

4. 4

   Distinguish untrained from unkind — and act differently on each

   Untrained is workable: you can supply structure, drafts, and patience, and it usually improves. Unkind is not a skills gap, doesn't respond to better inputs, and the CMI attrition data says leaving is a rational and extremely common response. Deciding which one you have is the most important call in this piece.

5. 5

   If you're the accidental manager, say the quiet part once

   "I haven't done this before and I'm working it out" said deliberately, once, to your team, costs less authority than you think and buys more than almost anything else available. You are in the 82%. Your team has probably already worked that out, and the only thing your silence is protecting is your own discomfort.

Day five¶

I've thought a lot about how that conversation would have gone if she'd bluffed.

She'd have produced a decision. It would have sounded confident, because that's the register the role demands, and I'd have implemented it without much scrutiny because it came with the weight of the chair behind it. It would have been a guess wearing a suit, and I'd have had no way to tell.

Instead she said I don't know, what do you think, and we made a considerably better decision together in about twenty minutes — and I spent the following eight years arriving at those conversations with a draft instead of a question, which is the most valuable procedural thing anyone has ever accidentally taught me.

They're guessing. Four out of five of them will tell you so if you make it safe. The guess is not the problem — a guess with a document in front of it is just a decision with the working shown.

The problem was only ever that we've built a job where saying it out loud costs something.

It's just business.