How to type your boss is a three-week observation exercise, not a naming ritual. You can read the working style without ever naming a type, and the naming is where this usually goes wrong. The useful output is a working hypothesis: four observable dimensions, three observations each, one sentence you can test. Not a verdict about who they are.
So this is not a typing guide with the serial numbers filed off. It is a guide to noticing four things — how fast they close a decision, whether they want the conclusion or the background first, what they do in the first thirty seconds of a disagreement, and how far ahead they are looking — then using those to predict friction, and testing the prediction instead of believing it.
Why a verdict is worse than a hypothesis
Give someone a type and you stop collecting evidence. That is the whole mechanism. Once you have decided your manager is a certain kind of person, you notice the behaviour that confirms it and file the rest as exceptions, and the more senior they are, the less data you get for correcting yourself.
The second cost is that it turns a management problem into a personality problem. If the real issue is that nobody has told you what done means, or that the team is short-staffed and your boss is absorbing it, a type story reframes that as a mismatch between their character and yours. Character is the one thing neither of you can change on a Tuesday.
Four dimensions you can actually observe
Each is a behaviour, not an essence. You are not looking for evidence of a cognitive function; you are looking for what changes when the pressure changes. Nor are these the four letters reassembled — the sixteen types come from a stack of functions, and a stack is not recoverable from four habits.
- Decision speed. How fast do they close? The more useful version of the question: what do they need before closing — one number, one dissenting voice, one more week?
- Information order. Do they ask for the conclusion first and the reasoning after, or do they want the background before they will hear a recommendation?
- First move in conflict. In the first thirty seconds of a disagreement, do they press, pause, repair, or change the subject? The first move says more than the argument that follows it.
- Time horizon. When they justify a decision, do they reach for this quarter or for three years out? Listen to the tense they use when giving reasons, not the words on the slide.
Three observations, not one
One data point is an anecdote about a Tuesday. You want the same dimension seen three times, in three different scenes, because the scene is a confound: a boss under a board deadline looks decisive and impatient regardless of what they are like in a normal month.
- Routine. The weekly one-to-one or the standing review, the scene where nobody is under threat.
- Fire. Something broke, a client is angry, a number was missed. Watch the first move, not the recovery.
- Planning. The meeting where next quarter or next year gets decided. This is where the time horizon becomes audible.
What to write down
Five lines per observation, in a note nobody will ever see. The discipline is the difference between a hypothesis and a vibe.
- Date and scene, in four words.
- What they asked for first.
- How long the decision took, and what unlocked it.
- Their first move when you disagreed.
- The span of time they used to justify it.
Turning the hypothesis into better collaboration
The point of the hypothesis is a change you make, not a label you hold. Three steps, and the third is the one people skip. If most of your contact is written rather than spoken, the async piece runs the same logic at team scale.
- Change one input and watch. Lead with the number instead of the story for two weeks. If decisions get faster, your hypothesis about information order was useful, whether or not it was true.
- Test the opposite reading. Deliberately try the account that is least flattering to your theory: not impatient but overloaded, not vague but uninterested in this particular decision. If the opposite predicts their behaviour just as well, you have a story rather than a hypothesis.
- Report the pattern, not the type. "You seem to decide faster when the number is in the first line" is a sentence a boss can use. "You are such a J about this" is a sentence that ends the conversation.
When the hypothesis is wrong
Most of the time it will be partly wrong, and the reasons are boring: role, incentives, stress, and the quarter they are having. A director in a cost-cutting cycle behaves like someone you would describe very differently in a growth year. Type is a claim about preference, not a prediction of behaviour under load, and the piece on type under stress is the honest version of that limit.
FAQ
Can I read a boss's type from one meeting?
No, and not from one good meeting either. A single encounter shows you their behaviour in a scene with a purpose: hiring you, selling you something, closing a quarter. Three observations in three different scenes is the minimum before a pattern is worth writing down, and even then it is a hypothesis about preferences, not a diagnosis.
Isn't this just stereotyping with extra steps?
Close enough that it is worth answering directly. The difference is falsifiability and scope: these four dimensions are about behaviour you have watched, they come with a prediction you can check in two weeks, and they are supposed to be discarded when the prediction fails. A stereotype is a claim about a category that survives every counterexample.
What if my boss tells me their type?
Treat it as a hypothesis they handed you, and one they may hold tightly. The useful question is not whether the label fits but what it predicts: how they want bad news delivered, how much context they want before a decision, how far ahead they are planning. Then watch. People are often wrong about themselves in exactly the way that matters at work.
Does any of this help if the boss is the problem?
Sometimes no amount of reframing fixes a badly designed job or a manager who does not defend their team. Observations help there too: they give you specific, datable examples instead of an impression, which is what you need before a conversation with your boss's boss. If the pattern is that nothing you change changes anything, the finding is about the arrangement, not the personality.
Written from published type theory, community reports and observable patterns — not from a clinical sample. Framework language, not diagnosis. Updated November 2026.