What I Learned About Board Governance from Sitting Across From a Public Company Board

Second essay in the “What I Learned From Sitting Across From…” series. As Chairman and CEO of a public mining company, I sat on both sides of the governance table at the same time. That combination is uncommon enough to be worth writing about, because it teaches you something specific that neither role alone can. What the board is actually for is the answer most founders only learn when it is too late to install it. · Juan Vegarra

Almost no founder understands what a board is actually for until they have been on the other side of a public company board that had real fiduciary duty, real reputational risk, and real willingness to fire the CEO. Startup boards do not function this way. Founder-controlled boards do not function this way. Boards with ceremonial independent directors do not function this way. Real boards function differently, and the difference only becomes visible when you have watched one operate up close.

For most of my time at Vena Resources, I held the roles of Chairman and CEO simultaneously. Combining those roles is unusual in a public company and increasingly discouraged by governance advocates, for reasons I understand better now than I did then. It is also a specific vantage point on what a board actually does. From the CEO seat you see what the board demands of you. From the Chairman seat you see what the board demands of itself. Doing both at the same time makes both visible in a way that neither role alone could reveal.

What the board is actually for

The board is not for endorsing your strategy. It is not for helping you raise capital. It is not for opening doors to customers. Those things can be legitimate secondary benefits of good directors, but they are not the reason the board exists. The board exists to protect shareholders from the CEO, including from the CEO’s own optimism. That is a specific job, and it is a job that founder-controlled boards structurally cannot do.

The Uber board in 2017 provides the canonical example of what a real board does when it decides it has to. In June 2017, five of Uber’s most prominent investors, including Benchmark, delivered a letter titled “Moving Uber Forward” that demanded CEO Travis Kalanick step aside.^[1] Kalanick resigned. Two months later, in August 2017, Benchmark filed suit in Delaware Chancery Court alleging that Kalanick had fraudulently obtained approval to expand the board with allies who would keep him in a position of influence.^[2] By October 2017, following the SoftBank $9.3 billion investment that closed in January 2018, Uber’s board voted to eliminate the 10x supervoting share structure that had given Kalanick and Benchmark disproportionate voting influence.^[3]

What that sequence demonstrated is that a real board can and will remove a founder-CEO even at the world’s highest-valued private company. That capacity is what gives the board its role. Startups whose boards do not have that capacity are startups whose boards are not doing the job the board exists to do. Whether the capacity ever gets exercised is a different question from whether it exists.

What the CEO seat teaches you about being governed

Sitting in the CEO chair while a real board governs you teaches a specific discipline that founders raised on their own controlled boards never develop. You learn what it feels like to have every material decision reviewed by people whose job is not to agree with you. You learn to build the paper trail that supports the decision before you announce the decision, rather than after. You learn to walk into every executive session assuming three of the seven directors will read the materials carefully and the other four will scan them, and to design the materials so both audiences reach the right conclusion.

You also learn what it feels like to walk into an executive session you know is about you. That is a moment founder-controlled CEOs rarely experience. When it happens, the CEO’s prior six months of behavior determines the outcome. If the CEO has been rigorous about disclosure, thorough about escalation, honest about risks that have not yet materialized, and consistent in the pattern of how decisions are made, the executive session is a working conversation. If the CEO has been selective about disclosure, defensive about risks, or inconsistent in how decisions get presented, the executive session is a different kind of conversation. There is no way to reconstruct rigor retroactively at that moment. The prior six months are what they are.

The founder-CEO who never experiences a board that could genuinely remove them never learns this discipline. And the CEO who never learns the discipline builds a company that eventually pays the price for it, usually at a moment of financial or reputational stress when the discipline would have made the difference between survival and collapse.

What the Chairman seat teaches you about running a board

Running the board from the Chairman seat teaches an entirely different set of skills. The Chairman is responsible for making the board effective as a governance body, which means allocating agenda time correctly, managing the balance between operational reporting and strategic discussion, ensuring committees do their actual work, and recognizing when the board itself is drifting into either rubber-stamping or micromanagement.

The specific hardest thing about the Chairman’s job is calibrating executive session time. Executive sessions are the periods when the board meets without management present. Too little executive session time means the CEO controls what the board discusses. Too much executive session time means the board loses touch with the actual operation of the company. The calibration between those two failure modes is a Chairman skill that only develops with practice, and it is where a Chairman-CEO combination gets structurally awkward. When I ran executive sessions at Vena about the CEO’s performance, I was running the sessions about myself.

That combination is exactly why current governance guidance strongly discourages combining Chairman and CEO roles in public companies. The independence problem is real. The Conference Board’s 2026 board-leadership report found that combined CEO-and-Chair roles at S&P 500 companies fell from 47 percent in 2020 to 42 percent in 2025, while independent chairs rose from 33 percent to 39 percent over the same period.^[5] Globally the trend is stronger. OECD data shows that 76 percent of jurisdictions with one-tier boards now require or encourage separation of the two roles, up from 44 percent in 2014.^[6] The direction of governance opinion is settled. What the combination does teach, though, is a visceral understanding of what each side of the table needs from the other, which becomes useful when you eventually sit in only one of the two chairs.

What happens when the board does not do its job

The counterexamples are more famous than the examples. The Theranos board included former Secretaries of State Henry Kissinger and George Shultz, former Senators Sam Nunn and Bill Frist, former Defense Secretary William Perry, and former Wells Fargo CEO Richard Kovacevich.^[4] That is one of the most distinguished boards ever assembled in Silicon Valley history. The company was valued at $9 billion at its peak. The technology did not work. The board never asked the questions that would have surfaced that fact, because the board had been assembled for prestige rather than for governance expertise. None of the directors had relevant healthcare, laboratory science, or diagnostics background. Prestige is not oversight. It is decoration.

The WeWork board had similar characteristics. Adam Neumann held supervoting shares with roughly 20 times the vote of common shares, could effectively appoint and remove directors at will, and had structured his ownership so that any external oversight was ceremonial.^[5] The board did not fail to prevent the S-1 disaster. The board was structurally incapable of preventing it. That distinction matters. A board that has been designed to be ceremonial is not going to become substantive at the moment substance is required. The design decision that mattered was made years earlier, at the moment the supervoting structure was accepted.

Every founder eventually decides how much governance capacity to build into their board. The decision looks trivial when the company is small. It becomes irreversible as the company scales. Founders who accept supervoting structures, board seats filled entirely with allies, and independent directors chosen for prestige rather than domain expertise are making a decision whose cost only shows up when the company faces the situation the board was supposed to be there to handle.

What founders should actually ask about a board

The specific test I would apply to any founder’s board is this. If the CEO had to be removed for cause tomorrow, could this board do it? If the answer is no, the board is not really a board. It is a group of people who advise the CEO. That may be useful. It is not governance.

The follow-up test is whether the board has the domain expertise to know when removal for cause is warranted. A board of five distinguished generalists cannot evaluate the technical claims of a diagnostics company. A board of five interventional cardiologists cannot evaluate the working capital dynamics of a hardware business. A board of five successful CEOs cannot substitute for a director who has actually served on an audit committee under securities laws. The board’s composition determines what the board can actually govern. Prestige without expertise is not neutral. It actively obscures the governance failure.

Founders building for the long run should insist on boards that could remove them, staffed with directors whose expertise matches the risks the company actually runs. That is a harder board to assemble than a friendly one, and it is a more useful board to have when the company faces the situation the board exists to handle.

What I would tell my earlier self

If I could go back to the moment I combined Chairman and CEO at Vena, I would separate the roles earlier and appoint a strong independent Chairman. The combination is defensible under certain conditions and was defensible in my case at the time. It is not the arrangement I would recommend to anyone starting from scratch today. The independence problem is real, and the specific value of a separated Chairman is greater than I appreciated when I held both roles.

The other thing I would tell my earlier self is that a strong board is not a threat to a strong CEO. The specific value of a real board is that it forces the CEO to develop the discipline that a founder-controlled board never demands. That discipline compounds. It is what makes the CEO better at the CEO’s job. Founders who fight to keep the board weak are optimizing for short-term autonomy at the cost of long-term capability. It is a bad trade, and the CEOs who make it are the ones most likely to end up in the executive session they cannot walk out of.

That is what I learned from sitting on both sides of the board table for years. It is not what founders raised on controlled boards want to hear. It is what founders raised on controlled boards eventually discover, usually at a moment when the discovery has become expensive.

Juan Vegarra is the author of An Outsider’s Playbook (2026). He held the combined Chairman and CEO role at Vena Resources during its listing on the Toronto Stock Exchange and subsequent operating period. More essays · Free toolkits · Advisory · Write me