The Truth Stays Downstairs

Thoughts from the treadmill¹

A company is, at its core, a shared story. People agree to act as if the org chart, the mission, and the quarterly plan are real, and that agreement is what lets them build together. It does not take thousands of people for this to take over. Walk into a hundred-person startup and the founders already don't code. The CEO and CTO haven't opened GitHub in a year.

What do they do instead? Not coordination. Watch closely and the actual cross-team coordination runs one or two levels down, carried by senior engineers and managers who move across streams and stitch the work together. The top job is something else: pulling everyone into a room, weekly or even daily, and selling the story. Vision, culture, inspiration, capital. It is an evangelist's job, and the diary study of 1,114 CEOs confirms it: the higher-performing CEO type runs multi-function, high-level gatherings, while the lower-performing type does one-on-one operational meetings².

Let me be fair to the people who hold that job, and precise about who gets it. The role selects for a surface: warmth, confidence, composure under fire, the ability to fill a room and make people believe. Very different inner architectures produce that surface. One is the genuinely caring stoic: leaders with deep empathy and trained calm, who keep functioning under extreme duress because they have practiced regulating what they very much do feel, and who hold large teams together because they care. Investors who trust such people are buying something real. The other is glibness riding on fearlessness. Clinical psychology even has a name for the calm half of that profile: boldness, "a capacity to remain calm and focused in situations involving pressure or threat", and it is a core dimension of psychopathy¹³. Read that definition twice. It is also every board's ideal chief executive. Measurable psychopathic traits sit somewhere between 4 and 16% of senior managers depending on where the threshold is drawn, against roughly 1% in the general population⁸, and the same research finds those traits score high on charisma and presentation while scoring low on actual responsibility and performance⁷. Psychopathy is a continuum rather than a category, so any percentage is a line on a sliding scale⁹. And beyond the measurable band there is the rest: power itself dulls empathy in ordinary people¹⁰, so part of the not-caring at altitude is manufactured by the chair rather than imported by its occupant. I will admit my own vantage point could bias me here; I watch this from the engineering manager floor. What is certain is that the selection filter cannot tell deep empathy from polished glibness, or trained calm from constitutional numbness, because it only ever tests the surface. So the top of organizations fills with both: the deeply empathetic and the deeply glib, indistinguishable in the boardroom, opposites on the ground.

The blind spot is measurable at every level. In small groups, the strongest predictor of who emerges as leader is speaking time, not intelligence, not task proficiency, not the quality of what is said³. In organizations, firms systematically promote their most visible performers even when the data says they will be worse managers⁴. At the top, half of founders are replaced within three years as the game shifts to distribution⁵, and the ones who survive are those whose technical knowledge is too deep to transfer⁶.

And the blind spot compounds, because of how truth travels. By fifty or a hundred people, functional roles have formed, and ground truth no longer moves upward as truth. It moves as KPIs, dashboards, and status reports, sanded a little smoother in every meeting room it passes through. By the time it reaches the top, the story is being told back to the storytellers.

And then comes the final turn of the screw: the storyteller starts believing it. Power inflates confidence in one's own judgment¹¹, and overconfident chief executives measurably destroy value, overpaying for deals the market immediately discounts¹². Watch today's AI scene and the loop runs in public: capabilities overrated to evangelize capital and talent, while the same inflated story confuses buyers, scares the workforce, and freezes decisions across whole industries. Evangelism unanchored from ground truth does not accelerate the future. It anesthetizes the present.

Dashboards will not fix that. People fix that. You need people at higher levels of management who can speak the ground truth and survive speaking it. The Romans solved this problem twenty-five centuries ago: when the plebeians seceded, Rome created the tribune of the plebs, an office whose holder was sacrosanct and could veto the patricians, precisely so the people doing the work had a protected voice at the top. We have built inclusion programs for nearly every dimension of identity. We have not built this one: the inclusion of people who know the ground, in the rooms where the story gets written.

You can tell which organizations have it without reading a single org chart. Look at the product. Look at the timelines. It reflects there, even in a ten-person startup.

The common objection writes itself: many capable technology leaders will say you do not need to do the work to understand it. They are half right, and I will not pretend otherwise; nobody is asking the CTO to ship production code. But understanding is not a fact you can be briefed on. It is a perspective, and some perspectives only open up from inside the act of building. The friction of the current toolchain, where it fights you, what it makes cheap and what it quietly makes impossible, is knowledge that lives in the hands before it can be put into words. A leader who last built something a decade ago is not merely working from stale data; they are missing a vantage point entirely, and no dashboard reconstructs it. We like to assume someone below will translate that view upward, but the most important parts are the hardest to translate, and often there is no one positioned to do it. So the leader signs off estimates, staffing, and AI strategy from a place they can no longer see. One uncomfortable stretch with the real tools does not refresh a number. It restores a line of sight.

There are engineers who inspire and evangelize ideas every day, hold their calm under pressure, and even close a sales pitch. The capability is not missing. The selection signal is. We pick leaders by who talks, sells, and fills the room, and we never test whether they still understand the work. The people skills are real, and the blind spot is real. The organizations that thrive will be the ones that stop leaving the truth-tellers downstairs.

Notes & References

1. Refined with Claude.
2. Bandiera, O., Hansen, S., Prat, A., & Sadun, R. (2020). CEO Behavior and Firm Performance. Journal of Political Economy, 128(4).
3. MacLaren, N. G., Yammarino, F. J., et al. (2020). Testing the babble hypothesis: Speaking time predicts leader emergence in small groups. The Leadership Quarterly.
4. Benson, A., Li, D., & Shue, K. (2019). Promotions and the Peter Principle. The Quarterly Journal of Economics, 134(4).
5. Wasserman, N. (2008). The Founder's Dilemma. Harvard Business Review.
6. Technical Specialized Knowledge and Founder Leadership at Initial Public Offering (study of 2,000+ IPOs).
7. Babiak, P., Neumann, C. S., & Hare, R. D. (2010). Corporate Psychopathy: Talking the Walk. Behavioral Sciences & the Law.
8. Boddy, C. R., et al. (2011), prevalence range documented in: Corporate law and corporate psychopaths (review).
9. Edens, J. F., Marcus, D. K., Lilienfeld, S. O., & Poythress, N. G. (2006). Psychopathic, Not Psychopath: Taxometric Evidence for the Dimensional Structure of Psychopathy. Journal of Abnormal Psychology, 115(1).
10. Hogeveen, J., Inzlicht, M., & Obhi, S. S. (2014). Power Changes How the Brain Responds to Others. Journal of Experimental Psychology: General, 143(2). Coverage: NPR.
11. Power-led CEO overconfidence and M&A decision making (power as a source of CEO overconfidence, building on Fast et al., 2012). The North American Journal of Economics and Finance.
12. Malmendier, U., & Tate, G. (2008). Who Makes Acquisitions? CEO Overconfidence and the Market's Reaction. Journal of Financial Economics, 89(1).
13. Patrick, C. J., Fowles, D. C., & Krueger, R. F. (2009). Triarchic conceptualization of psychopathy. Development and Psychopathology, 21(3). Boldness definition discussed in: APA.