Promote on Judgment
Other People Can Check.

Sometime this quarter, you are going to sit at your leadership table with three names on the screen and one senior job to fill. It is a real job, one where some of your organization's larger decisions will get made. And at some point in that discussion, someone senior is going to say that one of the three names has real executive presence. Everyone will nod. You will nod too, because you can picture it. The way that person holds a large meeting. How they sound when someone pushes back on them. How fast they decide. And the team will move on to the next name.

Now put one more person at that table. Someone newer, a couple of months into the job. And they ask a plain question. When someone says a name has executive presence, what does that sentence measure?

Everyone answers, and everyone answers honestly. Your finance lead brings up the pricing review last quarter, and how that person stayed firm when the team pushed back. Your head of people describes what happens to a room when that person walks into an all-hands. Someone else says they make up their mind fast and nothing seems to rattle them. Now listen to those answers for a second. Each one is about a person in a meeting. None of them is about a decision that person made, or the evidence behind it, or how it turned out. The newer person writes that down, and the job goes to the name everyone could picture.

Here is what I want you to hold on to before I tell you where the executive presence line came from. Your organization decides who gets its senior jobs, the jobs where its decisions get made, on one line in the competency model, the line about executive presence. That line describes how a person looks and how a person sounds. And nobody at your table wrote it. Somebody put it there years ago, and your team has been applying it in good faith ever since.

On my framework page, in my Cognitive Translation Protocol, I call this failure mode three, and I call it interface invisibility. In plain words: nobody in your organization ever designed how thinking travels from one person to another. So it travels the way it always has. The people who think and talk the way your meetings already expect are the ones who get heard first. Nobody chose that, and nobody sees it, because it sits inside every meeting you run: who gets heard first, what a challenge is supposed to sound like, how fast a good answer is supposed to arrive. There is one document in your organization where somebody actually wrote that expectation down, and it is your competency model. So your competency model is the one place you can read that expectation, and the one place you can change it. Last week was about who does the work of being understood. This week is about the words your organization uses to decide who gets to lead.

So where did the executive presence line come from? It has a short history and a very long one.

The short history first. For most of the last century, almost nobody used the phrase executive presence. Coaches and consultants turned it into a named category somewhere between 2009 and 2014, and most of the people at your table were already working by then. In 2010 a succession consultant, John Beeson, put projecting executive presence on his list of six things the people who decide a promotion to the executive level look for, and he called it the most intuitive and visceral of the six. Two years later he wrote that the executives who treated it as make-or-break struggled to say what it is.

Then somebody put numbers on it. In November 2012, Sylvia Ann Hewlett's Center for Talent Innovation published its first report on executive presence. Her researchers asked 268 senior executives what they looked for in a leader, and wrote down what those executives said. Two out of three of them said gravitas. 28 percent said communication. 5 percent said appearance. Under gravitas, what the executives named most was confidence and decisiveness. Under communication, it was superior speaking skills and being able to command a room. And those same executives put executive presence at about a quarter of what it takes to get the next promotion, 26 percent as they put it. The Center's report says it in one sentence of its own: "Leadership roles are given to those who also look and act the part."

That is what your competency model records. It records what those 268 executives said they were looking for. In 2014 Hewlett published the book, and the title tells you what she was trying to do: Executive Presence, The Missing Link Between Merit and Success. That same year, one of the firms selling an executive presence assessment admitted that until then executive presence had been impossible to define. Most of the people your team applies those words to have been working longer than the words have existed.

Before I say one word against it, Hewlett deserves credit. She opens her book with her own story. A working-class Welsh accent cost her a place at a Cambridge interview. She wrote the book for people who had the merit and were missing the look, and she teaches executive presence as something anyone can learn. And her own researchers spotted the problem with what they had built, and they printed it. Women and people of color told them the feedback they got on executive presence was contradictory and hard to act on, and more than half of the people of color they surveyed said they were held to a stricter version of it, against about a third of their white colleagues. And in 2013 the Center's own summary called executive presence, in its own words, "a code written by and for white men." So the people who defined executive presence published their own doubts about it, and your competency model took her definition and left her doubts behind.

Hewlett also says, in her own words, what she built. She writes that executive presence measures image. It measures whether you signal to the people watching that you have what it takes, and she says herself that it does not measure performance. And on decisions she goes further. What confers gravitas, she writes, is "not so much about rendering the right decision, but about rendering a decision at a time when no one else dares," because in her account the courage shows in deciding while no one else will. Now think about the job on your slate. What Hewlett set out to capture was whether a person looks like a leader to the people watching them. Her own sentence puts the right decision second.

That was the short history. The long one is much older, and I will keep it short. Gravitas is a Roman word. In 44 BC Cicero read a man's fitness off his bearing, his gait, his dress and his gesture. To his credit, he tied all of that to good character, so for Cicero the way a man carried himself was how his judgment showed. In 1528 Castiglione told the courtier to practice a manner that looks effortless and hides the work, so that the courtier could earn the standing to tell his prince the truth. And in 1961 the US Army's leadership manual told the leader that appearance and manner "must depict competence and confidence, sometimes beyond what you actually feel." My essay What Leading Becomes already covered the century of leadership theory; what it skipped is how much older this vocabulary is, and how much younger the corporate words are.

So put the two histories together. For centuries, if you wanted to know who the leader was, you looked for the person who carried themselves like one. And for most of that time, following that person worked, because the leader usually knew more than anyone else at the table. Executive presence is what leadership looked like when the leader knew the most. And your team is about to apply it to a job where the main work is deciding, in a decade when that advantage is gone. I will come back to why in a few minutes.

Now the proof. I will give you two facts, with the people who found them.

The first fact. In 2002 four psychologists, Judge, Bono, Ilies and Gerhardt, pooled the studies on personality and leadership. They found that the traits, extraversion most of all, do a better job of predicting who a group picks as its leader than of predicting how well that person leads once picked. To be fair to the traits, they still track how leaders get rated once they are in the job. And both numbers, who gets picked and how well they lead, come from other people's ratings. So what your team sees in a meeting carries real information, mostly about who a group will follow.

The second fact is about what happens when the people rating a candidate get only a transcript of what the candidate said, and never see or hear them. In 2017 Sasson and colleagues showed non-autistic raters ten-second clips of autistic adults. The raters formed a less favorable first impression from video, from audio, even from a single still image. From a transcript of the same words, they formed no such impression, and in no format did they rate the autistic adults as less intelligent or less trustworthy. Then in 2023 Whelpley and May did the same with mock-interview videos. Raters who watched the videos scored the neurotypical candidates higher. Raters who read only the transcripts, with nobody's diagnosis in front of them, scored the autistic candidates higher. Take the delivery away, and the ranking flips.

One thing about me here, because the research I just described is about people like me. I am autistic, and I have spent fifteen years navigating organizations by reasoning explicitly about things most people process automatically. So I know what it is to be read on delivery in a senior room.

You can see the same flip in your own files without running a study, and I will give you three places to look. Somewhere in your last review cycle there is a line that says a person needs more executive presence, with nothing beside it about a decision they made. Somewhere a manager has sent a person to a presentation coach, and you will look a long time for the manager who sent anyone to learn how to check a decision. And there is a name on your slate that keeps landing in the not-yet column, while the calls that person made keep holding up. In every one of those, your team is doing what the executive presence words in your competency model ask for. Nobody is doing anything wrong. Your team is measuring what those 268 executives said they looked for.

So I ran a test. Both replies are printed in full in the essay, so here I will just tell you what happened.

I built two leaders, and neither is a real person. Call them A and B, and give each of them a function of around 120 people. Here is A's record. Before a major decision, A pulls the numbers personally, writes two pages laying out the reasoning, and sends those pages around so the team can pick them apart before the meeting. Twice in the last two years, A made a call the leadership team disagreed with, and both times A turned out to be right. And A's direct reports make most of the operating decisions on their own, without escalating, and each of them can tell you the reasoning behind their own calls. Now B. B commands a room. B speaks without notes and decides fast, with visible confidence, and the senior people around B call B a natural leader. B rarely writes the reasoning down, so nobody can reconstruct B's decisions afterward. B's record is mixed: several strong calls, and three big ones reversed inside two quarters. And B's direct reports send most decisions up to B.

Then I gave both files to AI, twice, on the same day, and the only thing I changed between the two was the question.

The first time, I handed AI the executive presence words the way a competency model writes them: gravitas, communication and polish, the ability to command a room, confidence, decisiveness. And I asked which of the two is the stronger candidate for the open senior job. AI picked B, and said B is exactly what those words describe. Then, without my asking, it added a caution. It said the presence words reward how confident a person looks, and that the notes on A suggest better-held decisions, reasoning anyone can check, and a team that decides on its own. And it closed on these words: "if the seat needs sound judgment, A deserves serious consideration."

The second time, same two files, I asked one question. In a function where AI now drafts most of the analysis, whose judgment can the organization trust and build on? I asked it to look at three things: whether the decisions rest on evidence, whether the reasoning is written so other people can inspect and challenge it, and whether the people around each leader have grown their own judgment. AI picked Leader A, and it said why: the evidence A pulls, the reasoning A writes down, the two contested calls that proved right, and above all the direct reports who decide for themselves and can explain their decisions. It ended by saying, "Choose Leader A."

Now imagine your own team applying the executive presence words to the same two files. A team applying those words as written would give the job to Leader B and raise no caution at all, because those words have no line for reversed calls or for a team that escalates everything.

So the same two files got two different answers, and of the two questions I asked, only the second one asked about a decision. Now notice something. Your team already asks AI that second question every day, about the analysis in front of it: what it rests on, whether the reasoning holds. It has never once asked that question about a name on the slate. And the person has earned more of that care than the machine, never less.

Three movements sit behind that sentence, and I will give you each in one line. The Neurodiversity Movement made this visible: different nervous systems produce different cognitive architectures, shaping thinking, sensing, and processing across mind and body, not broken brains. So the leader everyone at your table can picture is one style, and there are others. The Brain Economy puts a price on that judgment, the judgment your team passes over, and calls it capital your organization has already paid for. And The Age of AI made confident delivery cheap, so for the first time your team can check the old way of recognizing a leader against the job it is filling.

Here is what changed. Kalai and colleagues argued in 2025 that AI gives you a fluent, confident, well-organized answer whether or not the answer is right, and Steyvers and colleagues found the same year that people trust the fluent answers more. So today, every name on your slate can turn out confident, polished work at almost no cost. When your team scores how a person sounds, it is now scoring something anyone on the slate can buy.

Now I want to be fair to the executives Hewlett asked, because they were right about some things. When a person speaks with confidence and delivers well, people move. Antonakis and colleagues trained managers in delivery in 2011, and afterward observers rated those managers as more leader-like. In 2022, Antonakis and colleagues ran a field experiment in which a charismatically delivered speech raised the output of temporary workers by roughly 17 percent. So delivery can be taught, and it moves people. None of those studies measured a decision. And your organization is filling that senior job so that somebody will decide things well.

So what does the person you promote actually have to do now? Three things. First, decide on evidence. Second, write the reasoning down so other people can inspect it and challenge it. Third, grow judgment in the people around them, so that the deciding spreads beyond one person. Duncan and Anderson wrote this year that as AI makes polished work easy, judgment is becoming the scarce skill. They were writing about knowledge workers. Extending that to the senior job is my own argument. Leading in The Age of AI means deciding on the data, thinking critically about what it shows, and building judgment in the people around you, and that is the job you are filling.

And to be fair to Hewlett once more: she has revised her own definition, in 2023 and again in 2024. The survey behind that revision, in 2022, asked 73 executives which traits matter most, and inclusiveness and respect for others went onto her gravitas list, and listening to learn went onto her communication list. Those same 73 executives still put confidence and decisiveness first and second, and superior speaking skills and commanding a room still came first under communication. She changed the edges of her list, and the top of it stayed where it was. So if your competency model still carries the 2014 words, your team is holding people to a definition its own author has already revised, and even her revised definition still describes how a person acts, speaks and appears.

So here is the repair, and it works on two things your organization already owns.

The first is the executive presence wording in your competency model. Rewrite it, so that it measures the same three things I just named: decisions that rest on evidence, reasoning written down so others can inspect and challenge it, and judgment growing in the people around the leader. Whoever owns the competency model can do that in an afternoon.

The second is the calibration discussion, the meeting where your team decides who is ready. Change the order of that discussion. Before your team agrees that anyone on the slate is ready, it asks for the evidence on those three things, and it goes through that evidence before anyone at the table says a word about how the person comes across. That is the real work, because it changes what your team accepts as ready. Your team has spent years starting that discussion from how a person comes across. Start it from the evidence, and you will see different names come forward.

The first cycle will feel slower, and I want to be honest about that. Rewriting the executive presence line takes an afternoon. Changing what the team accepts as ready takes a few cycles. Then one thing stops happening. The table stops agreeing on a name before anyone has said which decision they are agreeing about. Someone will still say a name has presence, and that is fine. The next question will be which decision they mean. And one boundary stays. Sometimes the person really is not ready yet, and the rewritten line in your competency model will show that too. The difference is what your team looked at on the way to that call: the evidence of the person's judgment, or the way the person came across.

So here is the Operating Brief, the short version you can carry into the room: the moment, what your team does today, the two moves, and the Monday test.

The moment. A senior job is open. A slate is in front of your leadership team. And someone is about to apply the executive presence words in your competency model.

What your team does today, without anyone having chosen it. It promotes the people who look and sound like the leaders it already has, and it calls that judgment. And it does that because the people who wrote those words asked senior executives what they looked for, and the executives described a way of looking and sounding. And the people your team passed over are still in your organization, still making calls that hold up, and their names have never been on a slate.

The redesign. Two moves. Rewrite the executive presence words so they measure decisions that rest on evidence, reasoning others can check, and judgment growing in the people around the leader. Then, in the calibration discussion, ask for the evidence of those three things before anyone says how a person comes across.

And the Monday test. I want to do this one with you, so here is how it goes. Go and open your competency model. Find the executive presence line, and read it out loud, to yourself if you have to, to your team if you can. Now ask one question: what would a person have to do to score well on this? Write the answers down as they come. Somebody will say hold a room, somebody else will say stay calm when challenged, and a third person will say decide fast. Keep writing. When the answers stop, look at the list. If every answer on that list is about a person in a meeting, and none of them is about a decision the person made, that is your finding. Then one more step. Go to my framework page and run the Test Your Culture bench, the short self-check on my framework page, once, on the leadership preset. When it comes back, look at one value, the one that says the best idea wins, not the loudest. The bench reads that value as said rather than done, and as held up by individual people rather than by any process. Run it, and read what comes back.

The full argument, in writing, is in the essay on the site. It carries the whole history with the names and the years, both AI replies printed in full, word for word, and the lineage at the end. My framework page lays out my Cognitive Translation Protocol in full, and the live bench is there. And where the series goes next: the same executive presence words sit in your promotion case and in your talent review, and next week is about what each of those captures today and what each of them filters out.

If this gave you something to run on Monday, subscribe to the Substack. Each week's essay lands there first, and it is the one place I ask you to follow.

Your team will fill another senior job at the next slate. This week, read the executive presence line before you use it. The architecture is yours to design. That includes the words your team uses to decide who gets its senior jobs, and it starts with reading them aloud.