Forex≈ 11 min read

Founder Mode: How Much Power Should One Person Have in a Company?

Founder Mode: How much power should one CEO have?

Founder mode can help a company maintain strategic focus, move faster and preserve the culture that made it successful. But it becomes dangerous when too much knowledge, authority and decision-making depend on one individual. Brian Chesky’s approach at Airbnb and Jensen Huang’s former leadership of Nvidia illustrate the potential advantages of highly involved CEOs: fewer organisational silos, direct communication and closer attention to execution. Research suggests, however, that founder-led and professionally managed companies often create value in different ways, while concentrated executive power can increase performance volatility. The most effective leadership model is therefore not one in which the CEO controls everything. It is one in which the leader stays close to the business, develops capable decision-makers and builds an organisation that can perform without depending on a single person.

  1. Founder Mode: Why Some CEOs Refuse to Step Back

The idea of founder mode has become one of Silicon Valley’s most discussed approaches to management. The term was popularised by venture capitalist and essayist Paul Graham in September 2024, following a speech by Brian Chesky, Airbnb’s co-founder and CEO.

Chesky challenged the conventional advice given to founders as their companies grow: hire talented managers, delegate responsibility and allow each department to operate independently. In his view, this approach can gradually distance the founder from the actual work of the business. Layers of management may filter information, create internal silos and weaken the organisation’s shared sense of purpose.

In his essay Founder Mode, published in September 2024, Graham described the distinction between two approaches to running a company. In manager mode, executives oversee their direct reports and generally avoid intervening in the details of lower-level teams. Founder mode, by contrast, allows the leader to engage directly with people and decisions several levels below the top of the organisational chart.

The distinction is not simply about how many decisions a CEO makes personally. It is about how information moves through an organisation, how priorities are communicated and whether employees understand the reasoning behind strategic choices.

Chesky has described his role at Airbnb as similar to that of an editor: reviewing work across the organisation, identifying inconsistencies and helping teams maintain a coherent direction. The approach is intended to keep the company’s strategy connected to execution rather than allowing departments to develop separate interpretations of corporate priorities.

This does not necessarily mean that every decision must pass through the founder. The stronger interpretation of founder mode is that the CEO remains deeply involved in the quality and direction of the work, even when responsibility for execution is delegated.

That distinction matters. A leader can stay close to a business without becoming the bottleneck for every operational decision.

  1. Jensen Huang and Nvidia: The Advantages of an Unusually Flat Organisation

Until recently, Jensen Huang was one of the most prominent examples of founder mode in practice. As Nvidia’s co-founder and longtime CEO, he became closely associated with the company’s strategy, engineering culture and position at the centre of the AI infrastructure boom.

Huang’s management structure was unusual for a company of Nvidia’s scale. He said that around 60 people reported directly to him and that he avoided conventional one-to-one meetings. Instead, he preferred group discussions in which multiple executives could hear the same information, examine problems together and understand how strategic decisions were reached.

In a 2024 conversation with Stripe CEO Patrick Collison, Huang argued that access to important information should not depend on an employee’s position in the hierarchy. He explained that he preferred communicating challenges, priorities and strategic direction to the wider leadership team at the same time.

This approach offered several potential advantages:

  • Faster information flow. Senior employees could hear the CEO’s priorities directly rather than relying entirely on several layers of reporting.
  • Shared context. Different teams could understand how their work related to the company’s broader objectives.
  • Collective problem-solving. Executives could challenge ideas and contribute expertise in a shared discussion.
  • Direct feedback. The CEO could observe how managers reasoned through problems instead of relying only on formal performance reports.

The approach also had a demanding side. A leader who frequently intervenes across organisational levels can create pressure, reduce managerial autonomy and make employees feel that their work is constantly under review.

More importantly, Nvidia’s success under Huang did not prove that this structure would work equally well in every company. The model was closely connected to his personal working style, technical understanding and ability to communicate with a large leadership group.

That dependence became a much more immediate question on November 12, 2026, when Nvidia announced that Huang had resigned as president, CEO and board member, effective immediately. The company’s filing with the SEC confirmed the departure and named CFO Colette Kress interim CEO while the board searches for a permanent successor.

The resignation followed reports concerning Huang’s personal relationships, including allegations involving a businesswoman suspected by US officials of links to Chinese intelligence. The Associated Press reported that it could not independently confirm some of the claims concerning a separate relationship with an Nvidia employee. Singapore’s investigation into the businesswoman and related companies concerned suspected false declarations involving server shipments; Singaporean authorities said they had not investigated her for espionage. These distinctions are important: allegations reported by the media should not be treated as proven findings.

The corporate governance question, however, extends beyond the circumstances of his departure. Nvidia now faces a practical test: can a company whose identity and strategic direction were so closely associated with its founder maintain its performance under a different leader?

  1. What Research Says About Founder CEOs and Professional Managers

The debate over founder mode is not limited to anecdotes from Silicon Valley. Academic research has examined what happens when companies move from founder leadership to professional management.

A 2022 working paper by Tristan J. Fitzgerald of Texas A&M University’s Mays Business School, ‘Til Death Do Us Part: The Relative Merits of Founder CEOs, examined companies in which founders had to be replaced following death or serious illness.

Fitzgerald used these unexpected leadership transitions as a natural experiment, comparing affected companies with similar businesses that continued to be led by their founders. The findings suggest that the two types of leadership bring different capabilities.

Companies that moved to professional CEOs experienced a 10% reduction in internally generated innovation relative to comparable founder-led firms. However, professional managers partly offset this decline through greater mergers and acquisitions activity, increased leverage and the development of larger, more stable top management teams. The combined effects on total firm value were broadly offsetting.

The implication is not that founders are inherently better leaders. Rather, founders and professional executives may create value through different mechanisms.

Founder CEOs can have a strong connection to the original product vision, greater willingness to pursue unconventional ideas and a deeper understanding of the company’s early culture. Professional managers may bring experience in organisational design, capital allocation, acquisitions and the coordination of complex operations.

A separate study by Renée Adams, Heitor Almeida and Daniel Ferreira examined the relationship between executive power and corporate performance. Their paper, initially circulated as an NYU working paper in 2001 and subsequently published in The Review of Financial Studies in 2005, found that stock returns were more variable at companies where CEOs held greater decision-making power.

This finding highlights an important trade-off. Concentrated authority can make decisions faster, but it also increases the potential impact of an individual executive’s judgement. When that judgement is correct, the company may benefit from decisive action. When it is wrong, fewer independent decision-makers may be able to challenge or correct it.

For investors, this matters because leadership structure can influence not only the direction of a business but also the variability of its financial performance.

  1. The Hidden Cost of Making One Person Indispensable

A company can benefit from a powerful and highly engaged CEO. But there is a difference between a leader who sets high standards and one whose presence becomes essential to every important process.

Excessive centralisation creates several organisational risks.

Decision-making bottlenecks. If too many approvals require the CEO’s involvement, routine decisions can slow down. Employees may spend more time waiting for instructions than solving problems.

Reduced managerial accountability. Managers who expect the founder to review or reverse their decisions may become reluctant to take responsibility. Delegation exists formally, but not in practice.

Information overload. A CEO who tries to monitor every department can become overwhelmed by operational detail. The more information flows directly to one person, the harder it becomes to distinguish strategic signals from routine noise.

Succession risk. If the company’s relationships, knowledge and decision-making processes are concentrated around one executive, their departure can create uncertainty across the organisation.

Weak internal challenge. Employees may hesitate to question a leader whose authority is rarely challenged. This can reduce the quality of strategic debate and make mistakes harder to identify.

These risks are not arguments for excessive bureaucracy. A large organisation with too many management layers can also become slow, politically fragmented and disconnected from customers.

The challenge is to create a structure in which senior leaders remain accessible without turning every issue into a decision for the CEO.

One practical approach is skip-level management: meetings between the CEO and employees several levels below them, without requiring every conversation to pass through the intervening management hierarchy.

The value of this practice is not that the CEO takes control away from middle managers. It is that both sides gain a more direct understanding of how the organisation works. Employees can communicate problems without excessive filtering, while the CEO can assess the quality of talent and execution beyond the immediate leadership team.

  1. The Real Test of Founder Mode Is What Happens After the Founder Leaves

The most revealing test of a leadership model comes when its central figure is no longer in charge.

A company that performs well only while its founder is present may have built a successful business, but it has not necessarily built a resilient institution. Its long-term strength depends on whether the founder has transferred knowledge, developed capable successors and established standards that can survive a change in leadership.

Apple and Disney illustrate how the influence of founders can persist long after their departure. Their founders’ original ideas, product philosophies and cultural standards became part of the companies’ identities. But preserving a founder’s legacy does not mean copying their management style indefinitely. It means translating the principles behind their success into systems that other leaders can maintain and develop.

For Nvidia, the transition following Huang’s resignation will provide an important test of this distinction. Colette Kress has taken over as interim CEO, while the board considers permanent candidates. The company’s challenge is to preserve its engineering culture, strategic focus and relationships with customers and partners without relying on the constant presence of the individual who helped build them.

For boards and investors, several questions are especially relevant:

  • Can the company maintain its strategic direction without the founder’s daily involvement?
  • Are key decisions understood and supported by more than one executive?
  • Do senior managers have the authority and experience to act independently?
  • Is critical knowledge distributed across teams rather than concentrated in one person?
  • Does the company have a credible succession plan?

These questions matter well beyond technology companies. In financial services, brokerage businesses and other organisations operating in fast-moving markets, leadership concentration can create operational and strategic vulnerabilities. A founder may understand the business better than anyone else, but that expertise becomes a corporate asset only when it can be shared, documented and transferred.

Conclusion

Founder mode is not inherently a better alternative to professional management. Its value lies in keeping leadership close to the product, employees and strategic decisions while avoiding the bureaucracy that can slow a growing company. But deep involvement must not become unlimited control. A strong CEO develops other decision-makers, encourages independent judgement and builds processes that remain effective when the leader is absent. Nvidia’s leadership transition brings this issue into sharp focus: the ultimate measure of a founder’s success is not how indispensable they become, but how well the organisation can perform after they leave.

By Ethan Blake