You’ve almost certainly done the preparation.
The material is ready. You know your opening. You’ve thought about how to frame it, and if it’s important you’ve rehearsed the first thirty seconds, because the first thirty seconds are where you either land or don’t.
Here’s the problem with all of that.
You prepared it before you knew who you’d be talking to.
Not their name and role — you know that. But what they need from you, what they’ll receive, what they’ll discount without noticing they’ve discounted it. You couldn’t know any of that in advance, and you’ve built your opening as though you could.
The read is available and it’s cheap. Here’s what to look at, in order of how reliable it is, and what to ignore.
What to Read
A person’s operating mode can be identified from a small number of high-reliability signals, most of which are available within the first minute of an exchange. The most informative is what they ask about first, since the initial question indicates the category of information they treat as decisive. Second is what they omit as self-evident, because people skip the layer they are fluent in. Third is sentence construction — whether conclusions are stated first or built toward — which indicates tolerance for process. Fourth is what they have arranged the encounter to optimise for, including duration, format, and whether documentation was requested in advance. Signals commonly relied upon but of low reliability include posture and gesture, appearance, and job title, all of which vary for reasons unrelated to how a person processes information. The practical consequence is that useful reading requires the other person to speak first, which conflicts with the standard practice of opening with prepared material.
Four signals worth reading, three that aren’t.
Signal One: What They Ask About First
The highest-value signal available, and it takes one question to obtain.
The first substantive question someone asks tells you what category of information they treat as decisive, because people ask about the thing that would settle it for them.
“How do we know that?” — evidence. They need the case before they’ll move.
“What does this get us?” — outcome. Consequence is the currency; method is noise.
“Who else is involved in this?” — relationship. They’re mapping the people before the proposal.
“What exactly do you mean by that?” — precision. Ambiguity blocks them; they can’t proceed while a term is doing two jobs.
One question, and you have a working read.
The four modes and what each needs from you are set out in the room where half your words disappear. What matters here is that the signal is this cheap and this early, and that almost nobody collects it because they’re busy delivering their opening.
Signal Two: What They Skip
The second most reliable, and it’s slightly subtler.
People omit the layer they’re fluent in. If something is obvious to you, you don’t mention it — you assume it’s shared.
So listen for what’s missing from how they describe a situation.
Someone who describes a problem without any reference to who’s affected is operating in a mode where people aren’t a variable. Someone who describes it entirely in terms of who wants what, with no mechanism, is operating in relationship. Someone who gives you the conclusion with no reasoning has fluency in the reasoning and considers it uninteresting.
This one’s useful because it works even when they haven’t asked you anything. You can read it from any two sentences of description.
Signal Three: How They Build a Sentence
Third, and it tells you about tolerance for process.
Conclusion first, then support: they think in outcomes and they’ll expect the same from you. Building slowly toward a point will lose them well before you arrive.
Building toward the conclusion: they value the reasoning and will distrust a conclusion that arrives without it. Give this person your answer first and they’ll feel something has been skipped.
You can read this from the first thirty seconds of anything they say, and it directly determines how you should structure your own delivery — mirror it. If they front-load, front-load. If they build, build.
Signal Four: What the Setup Tells You
Fourth, and it’s available before anyone speaks, which makes it the one you can use when you have no other data.
The encounter itself has been arranged by someone, and the arrangement carries information.
Duration. Fifteen minutes signals a decision-oriented mode. An hour signals exploration.
Whether a document was requested in advance. Someone who asks for pre-reading processes in writing and will want to spend the meeting on questions rather than exposition. Someone who didn’t want anything will want you to talk.
Who else is in the room. A person who brings two colleagues is either distributing a decision or creating witnesses, and both change what will work.
Whether there’s an agenda. Its presence indicates precision. Its absence indicates either seniority or relationship mode.
None of these is decisive alone. Together, before a word is spoken, they narrow the field considerably.
What Not to Read
Three things people rely on heavily that don’t carry the information they’re assumed to.
Body language. Crossed arms mean the room is cold roughly as often as they mean anything else. The popular repertoire of postural tells has a poor reliability record, and confidence in reading them is largely unearned. You’ll do better ignoring posture entirely than reading it badly, because a wrong read is worse than no read — it produces confident action in the wrong direction.
Appearance and manner. Both are heavily determined by industry convention, personal habit, and what was clean that morning. They tell you about norms, not about processing.
Job title. Titles indicate authority, not mode. Two people in identical roles frequently need completely opposite approaches, and assuming that seniority implies outcome-orientation will be wrong often enough to cost you.
The common feature of all three: they’re visible without the person having to say anything, which is exactly why they’re attractive and exactly why they’re weak. The reliable signals all require them to speak.
The Habit That Changes the Most
Which brings us to the single practical instruction in this piece, and it’s an uncomfortable one.
Don’t open with your content.
Open with a question, and spend the first ninety seconds reading.
Something cheap and genuine: “Before I get into it — what’s the part of this you most want addressed?” Or “What’s your read on where this currently sits?”
Then listen for signals one, two and three, and structure everything that follows accordingly.
This costs you ninety seconds and it costs you your prepared opening, which is the part you’re most confident in and most reluctant to give up.
What you get for it is a read, obtained before you commit, on the only variable that determines whether the rest of your material arrives intact.
Most people do the reverse. They deliver the prepared opening, watch it not land, and then attempt to adjust from a position they’ve already committed to — which is much harder than starting correctly.
When the Read Is Wrong
It will be sometimes, and the recovery is cheaper than people fear.
The tell that you’ve misread: no engagement rather than disagreement. Nodding, non-committal responses, nothing to grip. Disagreement means your content arrived and got assessed. Silence usually means it didn’t arrive.
When you notice, change register and continue. Don’t announce it, don’t apologise, don’t restart. Most listeners register a mid-conversation adjustment as attentiveness rather than as an error — and committing harder to a read that’s already failing is what actually costs you.
The full system this sits inside, with nine listener states rather than four, is Guiguzi’s nine types.
What This Is Not
It is not profiling. You’re reading a current state, not assigning a category. The same person will require different approaches from you on different days depending on circumstances, which is why a label is worse than useless — it stops you looking.
It is not manipulation. You’re changing how your argument is delivered, not what you’re arguing for. The test: would this still work if they knew you were doing it? Adapting your framing to a busy executive survives disclosure comfortably.
It is not a substitute for having something to say. Perfect delivery of an empty proposal is still an empty proposal. This determines whether good material arrives, not whether it’s good.
The Cost
The price is specific and it’s the reason most people won’t do it.
You have to give up the prepared opening.
That opening is the part you rehearsed. It’s where your preparation is most visible and where you feel most in control. Replacing it with a question and ninety seconds of listening feels like walking in underprepared, and it will feel that way every time, not just the first.
You’ll also, occasionally, ask the question and get a short unhelpful answer — “just take me through it” — which leaves you having spent your opening on nothing. That happens, and it’s survivable, and it’s the price of the times it works.
Second cost, longer-lasting: once you’re reading, you can’t stop. Every conversation becomes partly an assessment, including social ones you’d rather just be present in. A number of people find that more tiring than they expected.
Common Questions
How do you tell what kind of person you’re talking to?
Primarily from what they ask about first, since the initial question reveals what category of information would settle it for them. Secondarily from what they omit as obvious, how they construct sentences, and how they’ve arranged the encounter.
Can you read someone in the first minute?
You can read their current operating mode with useful accuracy in about two sentences. You can’t read their character or their competence that fast, and confusing the two is the main failure mode.
What about body language?
It’s much less reliable than it’s assumed to be, and a confident wrong read is worse than no read. Posture, appearance and job title all vary for reasons unrelated to how someone processes information.
What should I do before speaking?
Read the setup — duration, whether pre-reading was requested, who else is present, whether there’s an agenda. Then open with a question rather than your material, and spend ninety seconds listening before you commit.
What if I read them wrong?
Change register and continue without announcing it. Most listeners experience a mid-conversation adjustment as attentiveness. What’s costly is committing harder to a read that’s already visibly failing.
Isn’t this exhausting to do constantly?
It becomes largely automatic, though it doesn’t switch off, and some people find the permanent low-level assessment tiring. That’s a real cost worth knowing about in advance.
Where to Start
One meeting. This week. One change.
Replace your opening with a question.
Whatever you were going to say first, don’t. Instead ask something genuine and cheap — what part of this they most want addressed, or where they think it currently stands.
Then listen to the first thing they say and write one word in your notes: evidence, outcome, relationship, or precision.
Then lead with that.
Everything else about your material stays exactly the same. You’re not changing your argument, your position, or your recommendation. You’re changing which door it comes through, on the basis of ninety seconds of information you didn’t have before.
You’ll finish the meeting having used less of your preparation than usual and feeling slightly unmoored.
Check what happened to the decision.
I’ve spent twenty-five years working on how ordinary people develop capabilities that look impossible from the outside, and the person who walks into any room and is heard almost never has better material than everyone else.
They just spent the first ninety seconds finding out where to put it.