They Could Lose Anyone

Depth is a system output, not a talent output. The sports evidence proved it across countries, generations, and population sizes. The All Blacks sweep the sheds after every match, including World Cup finals, because the system is the thing that survives the players. The question is whether the same measurement works in business.

It does. And the results are just as clean.

The system is the product

Toyota

In 1949, Toyota's postwar demand collapsed. Production dropped from 992 vehicles in March to 304 in May. By the following spring, the company was days from bankruptcy. A bitter labour strike paralyzed the factory for two months. The founder's son, Kiichiro Toyoda, resigned in June 1950, taking responsibility for the layoffs that cut the workforce from 8,000 to 6,000. Two years later he died without seeing what his resignation made possible.¹

From that crisis, a production engineer named Taiichi Ohno began building what would become the Toyota Production System. The logic was simple. If a worker on the production line sees a defect, that worker pulls a cord and the entire line stops. Not a supervisor. Not a manager. The worker. The person closest to the problem has the authority to act on it. Nothing moves until the problem is solved. No inventory buffers to hide behind. No backlog to absorb the error. The system surfaces problems the moment they appear and forces them to be fixed before anything else happens.²

The system did not depend on Ohno. It did not depend on anyone. Toyota has had twelve presidents since its founding. Six from the Toyoda family. Six from outside it. Three consecutive non-family presidents ran the company from 1995 to 2009. Three consecutive family members ran it before them. The pattern is indifference, not alternation. The system does not care who is in charge. Akio Toyoda stepped down in April 2023. Koji Sato, no relation, took over. The system kept producing.³

In February 1997, a fire destroyed the Aisin Seiki factory that supplied 99 percent of Toyota's brake valves. Under just-in-time production, Toyota's factories held roughly four hours of stock. Analysts predicted weeks of shutdown. Thirty-six suppliers, supported by more than 150 subcontractors, self-organized within hours. They improvised fifty separate production lines for a part none of them normally manufactured. First replacement valves arrived in four days. Full production resumed within a week. No central directive coordinated the response. The supplier network had internalized the system so deeply that it could reorganize itself without being told to.⁴

In 2010, Toyota recalled nine million vehicles for unintended acceleration. Congressional hearings. Global media crisis. Akio Toyoda testified personally. The company accepted responsibility, enhanced its quality systems, and reclaimed its position as the world's largest automaker within two years. The error-correction mechanisms held.⁵

The product Toyota sells is cars. The product Toyota built is the production system. The cars are what the system produces. The system is what survives.

Like the All Blacks after losing a golden generation. Different players. Different decade. Same result. Because the result was never about the players.

Honda

Toyota proves a system can survive the people who built it. But surviving people is the easy version of the test. The harder question is whether the system can go somewhere it has never been.

Soichiro Honda founded his company in 1948 to build motorized bicycles. By 1949 he had produced the Dream D-Type, Honda's first full motorcycle. By 1959, Honda was the largest motorcycle manufacturer in the world. Then he did something unusual. He kept going.⁶

Cars in 1963. Power equipment. Marine engines in 1964. Bipedal robotics research beginning in 1986, culminating in ASIMO in 2000. The HondaJet received FAA certification in December 2015, twelve years after its prototype first flew. The engine mount configuration, which placed the engines above the wing rather than below it, broke an industry assumption so fundamental that no other business jet manufacturer had attempted it. The engineer who designed it, Michimasa Fujino, was a Honda engineer. Not an aerospace hire. A Honda engineer who applied Honda's problem-solving culture to a domain the company had never entered.⁷

Motorcycles, cars, jets, robots. Brazil produces both eleven-a-side footballers and futsal players from the same pipeline because the system produces capability, not category expertise. Honda produces both motorcycles and jets from the same engineering culture for the same reason.

Soichiro Honda made a pact with his co-founder, Takeo Fujisawa, that neither of their sons would enter the company. He retired voluntarily in 1973. The deliberate break from Japanese family-company norms forced institutional culture rather than personality-dependent culture. Honda's "waigaya" tradition, open debate sessions where rank dissolves and ideas are contested until proven valid or rejected, has survived eight leadership transitions across five decades. The system produces problem-solvers. What domain those problem-solvers enter next has never been the constraint.⁸

Action without philosophy is a lethal weapon; philosophy without action is worthless. Soichiro Honda

Honda said he thought best when he had a wrench in his hands. That culture did not leave when he did. It stayed because it was never about him. It was about the kind of person the system selected for, trained, and promoted. Every CEO since the founder has been an engineer. The system does not produce executives who happen to understand engineering. It produces engineers who happen to run the company.

Costco

Surviving leadership and entering new domains are both tests of what the system can produce. There is a third test. Whether the system can resist the pressure to become something else.

Jim Sinegal cofounded Costco in 1983 with a rule that functioned like a constitution: cap the markup at 14 percent on third-party products, 15 percent on Kirkland Signature. A typical grocery chain runs 25 to 30 percent. Walmart runs about 24 percent. Costco runs 12.⁹

In 2012, Craig Jelinek succeeded Sinegal as CEO. The cap held. In January 2024, Ron Vachris succeeded Jelinek. The cap held. Three CEOs across four decades. The same margin. The same $1.50 hot dog, unchanged since 1985. The same 93 percent membership renewal rate in the United States and Canada.¹⁰

The depth test for Costco is not whether the company can lose its CEO. It already has, twice. The test is whether the system can resist the one temptation that would destroy it. Every quarter, Costco could take 16 percent margins instead of 12. Nobody outside the company would notice. The earnings report would improve. Wall Street would applaud. And the wire connecting low prices to membership renewals to buying power to lower costs would sever. The flywheel would slow. The moat would drain. The empire would become the company that raised prices.¹¹

The margin cap is a structural constraint, like a salary cap in professional sports, not a business strategy in the way that phrase is normally used. It forces the organization to compete on efficiency rather than extraction. It forces every new CEO to inherit the same constraint the previous CEO operated under. It makes the system independent of the person running it. Sinegal is gone. Jelinek is gone. The cap remains. The renewal rate remains. The $270 billion in annual revenue remains.¹²

The membership fee is the profit. The merchandise is sold near cost. The low price is the reason people come back. The renewals are the earnings. Three CEOs understood this. The fourth will too, or the 92 percent renewal rate will tell them they got it wrong.

Amazon

Toyota survived transitions. Honda survived new domains. Costco survived temptation. The final version of the test is the most direct: can the system survive the loss of the person most identified with it?

Jeff Bezos stepped down as CEO on July 5, 2021, twenty-seven years to the day after he incorporated the company. Andy Jassy, who had joined Amazon in 1997 and built AWS from nothing, took over.¹³

In 2021, Amazon's revenue was $469.8 billion. In 2025, it was $716.9 billion. A 53 percent increase in four years under a different CEO. Operating income improved from $12.2 billion in 2022 to $36.9 billion in 2023, a 201 percent increase in a single year. AWS crossed $100 billion in annual revenue. The advertising business grew 24 percent year over year to $47 billion. The flywheel did not slow when the person who designed it left the building.¹⁴

Bezos described the mechanism in his 2001 letter: "Focus on cost improvement makes it possible for us to afford to lower prices, which drives growth. Growth spreads fixed costs across more sales, reducing cost per unit, which makes possible more price reductions. Please expect us to repeat this loop." They repeated the loop for twenty years under Bezos. They have repeated it for four years under Jassy.¹⁵

Amazon has entered e-commerce, cloud computing, logistics, streaming, advertising, grocery, healthcare, and satellite communications. AWS alone has cut prices at least 134 times since launching in 2006. The domain breadth is Honda's pattern: a system that produces capability across categories, not expertise in one. The difference is the transmission mechanism.

Honda transmits its culture through engineering selection and waigaya debate. Amazon transmits its through a phrase. "Day 1." Bezos wrote it in his first shareholder letter in 1997 and attached that letter to every subsequent one for the next twenty-three years. Amazon's Seattle headquarters building is named Day 1. When someone asked Bezos what Day 2 looks like, he answered: "Day 2 is stasis. Followed by irrelevance. Followed by excruciating, painful decline. Followed by death. And that is why it is always Day 1."¹⁶

The haka is performed before every All Blacks test match. It binds each squad to every squad that came before it. "Day 1" functions the same way. It is a cultural inheritance, repeated and ritualized until it becomes the thing the organization cannot forget, not a slogan.

The inverse

If depth were about people, losing the best ones would not matter as long as you hired replacements. It always matters. Because when the system changes, the output changes, regardless of how talented the people inside it are.

Boeing

In August 1997, Boeing acquired McDonnell Douglas for $13 billion. Boeing was the surviving company. McDonnell Douglas executives filled the key positions. Harry Stonecipher, the former McDonnell Douglas CEO, became Boeing's president and said it plainly: the intent was to run Boeing "like a business rather than a great engineering firm." In 2001, Boeing moved its headquarters from Seattle to Chicago. The stated goal was "greater separation" between leadership and the production plants. The engineers stayed in Seattle. The power moved to Chicago. Knowledge in one place, authority in another.¹⁷

The system changed. The output changed. 346 people died.

Boeing is the USA-in-soccer of aerospace. Same country. Same engineers. Same educational pipeline. Different system, different output. Before the merger, Boeing engineers had authority. A retired engineer described the old culture simply: "As an engineer, I had a lot of authority to do the right thing." After the merger, accountants made the decisions. The andon cord was not just missing. The structure was designed so that pulling it would end your career. Engineers who raised concerns about the 737 MAX were told not to rock the boat. A factory manager who recommended shutting down the production line was told: "The military's not a profit-making enterprise." A systems engineer who proposed a cross-checking sensor was overruled on cost.¹⁸

At Toyota, any worker can stop the line. At Boeing, the person who saw the problem was told not to mention it. Same industry. Same profession. Same country. One system surfaces errors. The other suppresses them. The engineers were just as capable. The structure determined what their capability could produce.

General Electric

In August 2000, General Electric was the most valuable company on Earth. Market capitalization: approximately $600 billion. Jack Welch was the peak. One of the most celebrated CEOs in American business history. He had built something enormous. He had not built something underneath it.¹⁹

Jeff Immelt took over in 2001. For the next sixteen years, GE reported profits every quarter. Revenue figures, margin beats, dividend payments. The narrative was intact. By the time Immelt left in 2017, GE's stock had fallen 43 percent while the Dow gained 121 percent. In 2018, GE was removed from the Dow Jones Industrial Average after 110 consecutive years. By 2024, the company was split into three pieces worth a combined $226 billion. Less than half of what Immelt inherited.

The depth test: remove the best person and see what remains. GE answered it. Welch left. What remained was a conglomerate with no system underneath the person who had held it together. No production system like Toyota's. No margin constitution like Costco's. No cultural transmission mechanism like Amazon's Day 1 or Honda's waigaya. Welch's discipline, "be number one or number two in every market or get out," was a personal standard, not a structural one. It left when he did.²⁰

Between 2015 and 2017, GE generated $30 billion from operations and spent $75 billion on buybacks, dividends, and acquisitions. The $45 billion gap was funded by borrowing and depleting reserves. The pension, examined after Immelt left, was underfunded by $31 billion. The company was profitable. The cash register was empty.²¹

GE under Welch is the peak that gets the headline. GE after Welch is the measurement nobody takes. The peak was real. The depth was not.

Kraft Heinz

In 2015, 3G Capital engineered the merger of Kraft and Heinz. They ran zero-based budgeting across the combined company. Nothing was sacred. Advertising dropped 39 percent below pre-merger levels. Research and development fell to 0.36 percent of sales, against a competitor average of 1.7 percent. Less than a quarter of what peers spent on the thing that determines whether a product improves or stagnates.²²

Earnings rose. Wall Street applauded. The cost-cutting was working.

Then consumers noticed. Store brands improved. Competitors kept investing. The Kraft Heinz brands that had survived on legacy recognition began losing shelf space. In February 2019, the company recorded a $15.4 billion write-down. The stock, which had traded at $97.77 in February 2017, fell below $30.²³

The mechanism is the same one that explains USA pay-to-play soccer. The American youth soccer system selects for family income before it selects for ability. It optimizes for the wrong input and produces an output that looks functional until it meets a system that optimized for the right one. 3G Capital optimized Kraft Heinz for cost extraction. It looked functional for years. The earnings improved. The margins expanded. The pipeline that produced the brands worth buying was starving the entire time. The system was consuming itself, and the quarterly earnings report could not detect the damage.²⁴

Boeing replaced an engineering system with a financial one. GE never had a system underneath the person. Kraft Heinz had a system and starved it. Three paths to the same destination. The depth was not there, and eventually the measurement caught up.

The question mark

Apple

Steve Jobs died on October 5, 2011. Tim Cook became CEO. In the fourteen years since, Apple has become the most valuable company on Earth. Revenue grew from $108 billion in fiscal 2011 to $416 billion in fiscal 2025. The stock has returned more than 1,000 percent. By every financial metric, the transition was extraordinary.²⁵

The depth test for Apple is category creation, not financial survival.

After Jobs returned to Apple in 1997, the company created the iPod (2001), the iPhone (2007), the iPad (2010), and the App Store (2008). Four new product categories in fourteen years. Each one redefined or created a market. Under Cook, Apple has created the Apple Watch (2015) and the Vision Pro (2024). The Watch succeeded. Vision Pro shipped at $3,499, sold an estimated 390,000 units in its first year, saw production reportedly halt by late 2024, and as of its second anniversary, Apple still had not decided what to do with it.²⁶

Between those two launches: nine years. AirPods (2016) and AirTag (2021) are accessories to the iPhone, not independent platforms. The HomePod was discontinued and relaunched. The product the system produces most reliably is a better iPhone.

Apple spends $34.5 billion a year on research and development. More than Honda spends. More than Toyota spends on R&D in most years. Cumulative R&D since 2013 exceeds $183 billion. The investment is enormous. The output, measured in new product categories rather than refinements of existing ones, has not matched the investment. iPhone remains 50 percent of total revenue. iPhone and Services together account for 76 percent.²⁷

This is not a failure. Apple under Cook has been spectacularly successful at what it does. The question is whether the system that produces extraordinary refinement also produces category creation. Whether the depth extends beyond making the thing it already makes better.

England's Premier League is the richest domestic football league in the world. Nine of the twenty highest-revenue clubs on earth play in England. The system produces the best league product on the planet. It does not produce the best national team. One major tournament win since 1966. Approximately 70 percent of Premier League players are foreign. The system optimizes for clubs, not country. The output follows the system.

Apple's system produces the best consumer technology refinement on earth. The M-series chips. The camera pipeline. The integration between hardware and software. Whether it produces the next platform, the way the iPhone was the next platform after the iPod, is a question the system has not yet answered. The R&D budget says it is trying. The product catalogue says the answer has not arrived.

The depth test does not ask whether a company can keep winning at the thing it already does. It asks whether the system can produce the next thing. Apple is the richest system in the world. The measurement is still open.

The measurement nobody takes

Business commentary measures peaks. Who reported the best quarter. Who had the highest stock price. Who made the cover. Peaks are dramatic. They fill earnings calls and conference keynotes. But peaks do not prove a system exists. A single great CEO can produce extraordinary results from anywhere.

Depth proves the system. And the test is the same one from sports: remove the best person and see what remains.

Toyota lost Ohno, lost Kiichiro, alternated between family and non-family presidents twelve times. The system kept producing. Honda lost its founder in 1973 and entered jets, robotics, and rockets without him. Costco lost Sinegal and the margin cap survived two successions. Amazon lost Bezos and revenue grew 53 percent in four years.

Boeing lost its engineering culture in a merger and 346 people died. GE lost Welch and $500 billion in shareholder value disappeared over sixteen years. Kraft Heinz starved its own pipeline and $15.4 billion vanished in a single write-down.

The pattern from sports holds precisely. The countries that passed the depth test had traceable systems: New Zealand's rugby pipeline, Canada's hockey conveyor belt, Brazil's futsal courts in every neighbourhood. The countries that failed had talent without structure: the United States in soccer, spending $8,000 a year per child and selecting for size over technique. The variable was never the people. It was always the system the people were fed into.

In business, the variable is the same. Toyota and Boeing employ engineers from the same universities. The engineers are not the difference. The structure that determines what their capability can produce is the difference. One system gives the person closest to the problem the authority to stop the line. The other tells the person closest to the problem not to rock the boat.

Peaks are visible. Systems are invisible. The peak gets the headline. The system is why there is always another peak behind it.

New Zealand lost McCaw, Carter, and the 2015 golden generation. The system produced the 2023 finalists. Toyota lost Ohno and produced the world's largest automaker. Amazon lost Bezos and the flywheel accelerated. Costco lost its founder and the hot dog still costs $1.50.

They have great people. Every company that passes the depth test has great people. The finding is that they could lose any of them, and the system would keep producing.

They could lose anyone. That is the measurement. That is the proof that the system, not the person, is the thing that was built.