You’ve had this happen.
You take an idea into a room. You’ve done the work. The logic is clean, the evidence is stacked, you’ve anticipated the objections and you’ve got answers for all of them.
And it dies.
Not violently. Nobody argues with you. There’s a nod, a “let’s pick this up,” and then it’s gone forever.
Two weeks later somebody with a worse version of the same idea, less prepared, slightly incoherent, gets it approved in eleven minutes.
You went home and told yourself it was politics. Maybe it partly was.
But here’s the thing worth noticing: it happens to you in specific rooms. Not all of them. There are rooms where you’re outstanding, where you can move anybody, where your instincts are just correct.
And there are rooms where half your words disappear on the way across the table.
Same you. Same preparation. Opposite result.
That’s not politics. That’s a matching problem, and somebody wrote down the solution to it about twenty-three hundred years ago.
What the Nine Types Are
Guiguzi’s nine types are a classification of people by conversational state, set out in the Warring States text attributed to Guiguzi, the “Master of Ghost Valley.” The core claim is that persuasion fails not because an argument is weak but because it is delivered in a register the listener cannot receive. The text pairs each type with the specific quality a speaker must lead with: the learned require breadth, the slow require clear distinctions, the argumentative require brevity, the high-status require leverage, the wealthy require elevation, the pressured require concrete benefit, the overlooked require deference, the bold require daring, and the person defending an error requires directness. The nine describe states a person is currently in rather than fixed personalities, which means the same individual moves between them depending on circumstance. The material sits in the chapter on weighing and appraisal, and it predates every modern personality framework by roughly two millennia.
That’s the system. Now let’s make it usable.
Where This Came From
Quick context, because the origin explains why the material is so unsentimental.
The Warring States period ran from roughly 475 to 221 BC. Seven kingdoms, permanently at each other’s throats, and a class of men who travelled between courts selling one commodity: the ability to change a ruler’s mind.
They had no armies. No territory. No hereditary claim. Their entire professional standing rested on walking into a throne room and coming out with a different policy than the one that was in place when they entered.
Get it right and you were given a chancellorship. Get it wrong and, depending on the ruler’s mood, you might not leave the building.
Two of the most famous — Su Qin and Zhang Yi — are recorded as students of Guiguzi. Su Qin is credited with holding six mutually hostile kingdoms in a coalition against the seventh. Zhang Yi is credited with taking that coalition apart. Same teacher. Opposite objectives. Both effective.
That’s the context in which the nine types were written down. Not as a theory of personality. As field notes from an environment where misreading a listener had immediate, physical consequences.
Which is why there’s no fluff in it.
The Turn: These Aren’t Personality Types
Here’s the thing you need before the list, or the list will be useless to you.
Every instinct you have will try to read these nine as categories of person. Filing cabinets. Nine drawers, put everyone in one, done.
That’s the wrong reading, and it’s exactly the mistake that makes modern personality frameworks so unreliable in live conversation.
These are states.
The same person is a different type on Tuesday than on Thursday. Your CFO is “the wealthy” — operating from surplus, uninterested in incremental value — right up until the quarter goes badly and he becomes “the pressured,” at which point everything that worked on him last month is now actively counterproductive.
Your most argumentative colleague, the one who contests everything, is not argumentative at home. He’s argumentative in the room where his standing depends on appearing rigorous.
This distinction is the entire difference between a framework that works and a framework that flatters you.
Type-thinking gives you a label, and a label gives you the pleasant feeling of understanding somebody, and then you say the same thing to them for six years while it stops working.
State-thinking makes you read the person in front of you today.
And it means you have to keep reading, permanently, which is more work — and is the reason most people prefer the filing cabinet.
The Nine
For each: the state, what it looks like, what you lead with, and what fails.
1. The learned — lead with breadth
The state: deep knowledge of a domain. They’ve seen the arguments before, including yours, in better versions.
What fails: depth. Going deeper into their field is how you get quietly filed as someone who doesn’t know what he doesn’t know. Also: preparing harder. Preparation inside their domain increases your exposure, it doesn’t reduce it.
What works: breadth. Bring the connection to an adjacent field, the pattern from a different industry, the thing outside their frame. You’re not competing on depth — you’d lose — you’re offering the one thing genuine expertise structurally lacks, which is peripheral vision.
Practically: stop trying to be a smaller version of them and be a different shape entirely.
2. The slow-following — lead with clear distinctions
The state: not stupid. Not following. There’s a difference and confusing them is fatal.
What fails: more detail, faster. Also “make it more impactful,” “you know what I mean,” and every other instruction that assumes shared context.
What works: distinction. This, not that. Boundaries, contrasts, one thing at a time. The mind that isn’t following doesn’t need more information, it needs edges around the information it already has.
The trap: explaining more clearly often means explaining more elaborately, which makes it worse. Clarity here means fewer moving parts, not better sentences.
3. The argumentative — lead with brevity
The state: contests everything. Plays devil’s advocate. Qualifies every statement. Leaves you exhausted.
What fails: more evidence. This is the counterintuitive one and it catches thorough people every time. Bring a bigger stack of proof to an argumentative listener and you’ve simply handed him more surface to attack. Every additional claim is another handhold.
What works: compression. The essential point, stated once, with nothing surplus attached. The fewer pieces you put on the board, the fewer moves he has.
A related and useful realisation: being right is not an argument, and treating it as one is why you keep losing exchanges you should win. That’s its own piece, and if you recognise yourself here, read it.
4. The high-status — lead with leverage
The state: senior, time-poor, thinks in outcomes. Fifteen minutes, and you’re the fourth thing on the list.
What fails: effort. Your process, your hours, the difficulty of the work. He does not care and it isn’t rudeness — effort is simply not the currency at that altitude.
What works: leverage. What this moves, what it prevents, what it makes possible, what the situation looks like afterwards. Consequence, not craft.
Also: thanking someone for their time at the top of a meeting quietly reframes the conversation as a favour granted rather than a matter of shared interest. Small thing. It costs you position and it’s completely avoidable.
5. The wealthy — lead with elevation
The state: operating from surplus. Money has stopped being the constraint, and therefore has stopped being persuasive.
What fails: value. Price-per-unit, durability, return on investment. Emphasising value to someone for whom money isn’t scarce reads as a category error — you’ve told him you think he’s someone else.
What works: elevation. Rarity, standard, meaning, the thing that isn’t purchasable at any price. Notice that luxury brands never mention how long the thing lasts. That isn’t an accident and it isn’t vanity. It’s this principle, running at industrial scale.
6. The pressured — lead with concrete benefit
The state: short-term trouble. Can’t think past this month. Financial stress, deadline, crisis.
What fails: long-term thinking. Strategy. Encouragement. A person in genuine short-term difficulty experiences “think about where you want to be in five years” as an insult, because it’s evidence you haven’t understood the actual situation.
What works: the specific, near, concrete benefit. What improves this week. Narrow the frame to match the frame they’re already in.
Additional mechanic worth knowing: giving someone under pressure a menu of options frequently makes things worse. Decision capacity is one of the first things scarcity consumes. One clear recommendation beats five good choices.
7. The overlooked — lead with deference
The state: the person nobody notices. Long-serving, invisible, no recognition in years. Often the person who actually knows how the building works.
What fails: generic praise. It lands as condescension because it’s obviously reflexive, and someone who’s been ignored for a decade has an extremely fine detector for reflexive.
What works: deference. Genuine. Ask them something only they’d know. Treat their answer as load-bearing.
And here’s the honest limit: this is the one you cannot fake. The other eight will tolerate a degree of technique. This one won’t, because a person who’s been unseen for years is calibrated specifically for detecting who’s performing. Attempt it insincerely and you’ll do more damage than ignoring them.
There’s a story in the tradition about a Chinese lord who learned more from a seventy-year-old gatekeeper than from his own advisers, and the reason the story survived two thousand years is that it’s counterintuitive and keeps being true.
8. The bold — lead with daring
The state: high energy, high risk appetite, moves before the analysis is complete.
What fails: caution. Warnings. Risk registers. To this person, caution reads as an assessment of him — you’re telling him he can’t handle it — and he will discount everything you say afterwards.
What works: the challenge. Aim the energy rather than trying to reduce it. “Here’s the hard version” gets further than “here’s why we should slow down.”
The boundary, and it’s a real one: only do this with someone whose confidence is actual. Dare a fragile person and you’ll get compliance followed by collapse, and that’s your doing. Telling the difference is a skill, and it matters more than the technique.
9. The one going wrong — lead with directness
The state: defending a decision that’s already failing. Committed publicly, now can’t reverse without cost.
What fails: gentleness. This is the most counterintuitive entry in the whole system and the most useful. Be soft with a person defending an error and you give them room to reinforce it — you’ve supplied a face-saving surface to lean on, and they will lean.
What works: sharpness. Said once, said clean, said without contempt.
The two conditions that make this legitimate rather than cruel: you say it once and don’t repeat it, and you stay afterwards. Repetition converts truth into noise the person can dismiss. Leaving converts honesty into an attack.
And there are situations where the right move is to say nothing at all — where the person cannot absorb it, or where you’re not the one who can say it. Knowing when not to deploy this is more valuable than deploying it well.
Why This Beats a Personality Test
Three hard lines, because this material gets bundled with three things it isn’t.
It is not MBTI, DISC, or any sorting instrument. Those assign a stable label from a questionnaire. This assigns nothing. It reads the state in front of you and adapts the delivery, then re-reads next time. The practical difference: a label tells you who someone is and is therefore wrong the moment circumstances change. A state read tells you what’s true right now, which is the only thing you can actually act on. Sorting frameworks also have a well-documented reliability problem — a meaningful share of people retest into a different category within weeks — and a system that reclassifies you between Tuesday and Thursday is describing a state while claiming to describe a type.
It is not manipulation. I’ll draw this line precisely rather than reassuringly. The difference isn’t the technique, because the technique is identical. The difference is whether the thing you’re moving someone toward serves them. Adapting your delivery so a busy executive can actually hear a proposal that’s good for his company is translation. Using the same read to move him toward something that costs him is fraud. The nine types are neutral, the way a lockpick is neutral, and anyone telling you the skill itself is clean or dirty hasn’t thought about it properly. There’s a full treatment in reading people versus manipulating them.
It is not body language. No micro-expressions, no crossed arms, no eye-direction. The signal here is structural: how someone builds a sentence, what they ask about first, what they treat as obvious. That’s considerably more reliable than posture, which is mostly noise dressed up as science.
The Cost
Every method on this site names its price. Here’s this one.
To use the nine types, you have to give up being right in the way you enjoy being right.
You’ve spent years building a case, marshalling evidence, constructing the airtight version. And this system tells you that with roughly a third of the people you’ll meet, the airtight version is the problem — that the argumentative listener needs less of it, the senior listener doesn’t want it, and the expert will read it as encroachment.
That’s genuinely unpleasant if your competence has been built on thoroughness. It feels like being asked to be worse at your job.
You’re not being asked to be worse. You’re being asked to stop optimising for a scoreboard nobody else is looking at.
The second cost is smaller and lasts longer: you can’t switch this off. Once you can see states, you see them constantly, in every conversation, including ones you’d rather just be present in. Reading people is a capability with no off switch, and a number of people find that tiring in a way they didn’t anticipate.
Where This Connects
The reason this material sits on the same site as a four-hundred-year-old method for changing your own fate isn’t decorative.
The whole argument starts with a man whose life was predicted accurately for twenty years, because he had a fixed character producing fixed output. He was readable. That’s the method for changing what was already decided.
The nine types are the same principle turned outward.
If a fixed character produces predictable output, then people are legible. Not mystically — mechanically. What Guiguzi’s students were doing in those throne rooms was reading the character and running it forward two moves, which is exactly what the fortune teller did to Yuan Huang.
One body of work is about becoming harder to predict. The other is about getting better at predicting. Same physics.
Common Questions
What are Guiguzi’s nine types of people?
Nine conversational states, each paired with what a speaker must lead with: the learned need breadth, the slow-following need clear distinctions, the argumentative need brevity, the high-status need leverage, the wealthy need elevation, the pressured need concrete benefit, the overlooked need deference, the bold need daring, and the person defending an error needs directness.
Who was Guiguzi?
A figure from the Warring States period known as the Master of Ghost Valley, traditionally named as the teacher of the strategists Su Qin and Zhang Yi. Whether he was one historical person, several, or a name attached to a school is genuinely unsettled, and the text bearing his name was likely compiled over time.
Can you really read someone in sixty seconds?
You can read their current state with useful accuracy in about two sentences, because sentence construction, what they ask about first, and what they treat as too obvious to mention are all high-signal. You cannot read their character that fast, and the failure mode is confusing the two.
Is this the same as personality typing?
No, and the difference is the point. Typing sorts people into stable categories. This reads a state that changes with circumstance. The same person will need three different approaches from you across a year.
Does this work if the person knows you’re doing it?
Mostly yes, which is a decent test of whether it’s manipulation. Explaining a proposal in terms of outcomes rather than effort doesn’t stop working when someone notices. Techniques that collapse under disclosure are usually the ones worth avoiding.
What if I misread someone mid-conversation?
Stop and change register. The recovery is much cheaper than people fear — most listeners register an adjustment as attentiveness rather than as an error. What’s expensive is committing harder to a read that’s already failing.
Where to Start
Not with all nine. Nine is too many to hold live, and you’ll end up thinking about the framework instead of the person.
Pick the room you fail in.
You already know which one it is. There’s a specific person or a specific meeting where you are consistently less effective than you are everywhere else, and where you’ve explained it to yourself as politics or personality clash.
This week, before you go in, answer one question: what state is this person actually in, and what am I currently leading with?
Nine times out of ten you’ll find you’re leading with the thing that works on you. Thorough people lead with evidence. Senior people lead with outcomes. Warm people lead with rapport. Everybody defaults to their own register and then experiences the mismatch as the other person being difficult.
Change what you lead with. Once. In one room.
That’s the whole exercise, and it’s usually enough to convince you the rest of the system is worth learning.
I’ve spent twenty-five years working on how ordinary people develop capabilities that look impossible from the outside, and the ability to read a room and adjust in real time is the one that most reliably gets described as uncanny by people watching it happen.
It isn’t uncanny. It’s a two-thousand-year-old checklist that almost nobody in the West has read.