The World’s Largest Bank with the Mindset of a Twelve-Person Startup
Jamie Dimon has been at the helm of JPMorgan Chase since 2006. Under his leadership, the bank has grown from $2.1 trillion to over $4 trillion in assets, absorbed Bear Stearns during the 2008 crisis, and established itself as the largest financial institution in the United States. In his annual letter to shareholders from April 2026, a 46-page document that serves as a managerial testament, Dimon did not speak of scale. Instead, he emphasized surgical precision.
His central thesis: the 320,000 employees of JPMorgan are the platform, but real competitive battles are won by small teams, focused on a single arena, whether it's a segment of investment banking, a specific type of client, or a product feature. He likened them to Navy SEALs or the Army's Delta Force: units empowered to make decisions and move with extreme speed, without waiting for approval from hierarchical layers that turn bureaucracy into the true enemy.
The metaphor is striking. Yet behind it lies an organizational diagnosis that few companies of that scale have the courage to publicly articulate.
What the Military Metaphor Reveals about the Real Problem
When a CEO of a company with over thirty years of corporate structure feels the need to invoke special operations forces to describe how he wants his teams to function, he is admitting something uncomfortable: that the organizational model he built over the years also generated exactly the bureaucracy he now seeks to dismantle.
This is not a critique of Dimon. It is an observation about the physics of large organizations. Approval layers, cross-validation processes, and alignment meetings do not appear by accident. They were built by leaders who, at some point, prioritized control over speed. Each new layer of oversight was, at one time, a rational response to an error, a regulatory scandal, or a loss of money. The accumulated result is what Dimon accurately calls the trench that needs to be broken.
Research on group dynamics, documented since Ringelmann's studies in the 19th century, confirms that individual effort decreases as team size increases. It is not a lack of commitment. It is social math. A team of five cannot dilute responsibility in the same way that a team of thirty can. Each member knows that their absence is visible. That visibility generates a type of pressure that no corporate performance evaluation system can artificially replicate.
Dimon is not discovering anything new. He is articulating aloud a tension that every mature organization has been managing quietly for years.
The Paradox of Scaling Without Losing Edge
What’s most interesting about Dimon’s stance is not just the bet on small teams. That is a common theme in any management book published in the last twenty years. What is interesting is that he is proposing this simultaneously with a massive integration of artificial intelligence into a workforce of 320,000, acknowledging that AI will eliminate some jobs while creating others, and committing to concrete plans for reskilling affected workers.
This combination of moves reveals the architecture of the real managerial problem: how to maintain the execution speed of a small team while managing the social and operational responsibilities of a corporation on a continental scale. There is no clean answer to that tension. There are options with distinct costs.
Meta did something similar. Mark Zuckerberg reorganized the company around small teams last year, describing that configuration as optimal for frontier research. The parallel is not coincidental. The largest companies in the world are converging on the same operational hypothesis: that competitive advantage doesn’t dwell in scale, but in the density of talent and autonomy within defined units. The startups currently competing with banks like JPMorgan, in payments or consumer credit, are growing by 20 to 30 percent annually with teams that are a fraction of 1% of their workforce.
The difference is not technological. It is structural. And that difference is what Dimon is trying to compress from within.
The Conversation That No Internal Memo Can Replace
There is a detail in Dimon’s letter that deserves more attention than the military metaphor. When he describes business platforms of AI, data, and financial systems, he insists that they must be highly efficient and that personnel should function like a well-oiled sports team. That phrase is not decorative. It is a diagnosis of what fails when corporate platforms become silos that produce reports but not decisions.
Bureaucracy is not a structural problem. It is the symptom of conversations that didn’t happen. Each redundant approval process exists because at some moment, someone avoided having the direct conversation about who actually has the authority to decide what. Each unnecessary alignment meeting is the institutionalized substitute for trust that was not built between the parties.
Simon Sinek, in his analyses of the special forces teams that Dimon references, points out something that corporate performance metrics systematically ignore: SEALs prefer the mid-performing member with high trust over the high performer with low trust. The latter, according to Sinek, is toxic in the long term. Research by Irrational Capital shows that teams with high interpersonal trust consistently outperform those organized solely by individual output metrics.
This means that Dimon’s bet on small teams only works if it is accompanied by a transformation in how JPMorgan defines merit, builds internal trust, and distributes decision-making authority without requiring upward validation at every step. A memo doesn’t achieve that. A letter to shareholders doesn’t either. It’s the accumulation of direct conversations that mid-level leaders choose to have or avoid each week.
The Model Dimon Proposes Comes at a Price Not Reflected in the Organizational Chart
JPMorgan's numbers are solid: $158 billion in revenue in 2025, with an expense-to-income ratio of 58% and a tangible equity return exceeding 15%. The investment banking sector holds about 25% of the global market share in merger and acquisition advisory. From that position, discussing SEAL-type teams sounds like strategic refinement on an already-won base.
But the risk lies precisely there. When an organization that is already winning begins to change its operational model in response to the pressure from nimbler competitors, the danger is not the change. It’s the speed at which the real culture adopts what the managerial rhetoric proclaims. The fintechs competing in payments and credit do not have 320,000 employees to relocate or regulatory structures to justify before the Federal Reserve. They have teams of thirty and a decision-making architecture that doesn’t require three levels of approval.
Dimon knows this. That’s why the language of the letter is that of someone diagnosing his own organization with the same coldness with which he would analyze a competitor. That ability to name the problem without softening it for the internal audience is, perhaps, the most challenging managerial skill to scale in an institution of that size.
The culture of any organization is the natural result of pursuing a purpose consistently over time, or the accumulated inventory of all the difficult conversations that institutional ego postponed until the market made them inevitable.
