Part IThe Story
The Caption Was Wrong in the Particulars
The photograph was taken on the floor of the New York Stock Exchange on Friday, April 28, 2017, by a Bloomberg photographer named Michael Nagle. It shows what such photographs always show: men in dark suits, the gilded trading floor behind them, the institutional cheerfulness of a listing day. The caption filed with it is the interesting part. It identifies "Ernie Garcia III, founder and chief executive officer of Carvana Co., second left," and beside him, at center, "his father Ernest Garcia II, chairman of Carvana Co."
The son was the chairman. He had taken the title at the IPO. The father held no formal role at the company at all. He was not an officer, not a director, not anyone whose name needed to appear in the governance section of the prospectus. The caption was wrong the way a rumor is wrong: in the particulars, not the substance. At the time of the offering, father and son together controlled 97 percent of Carvana's voting shares through various entities. Whoever stood at the center of the frame, the company belonged to the family.
The offering priced at $15 a share. Within the first week it was trading at $8. "We barely made it out," the son later told Car and Driver, which is not the sort of thing founders usually say about their IPOs. On another occasion he was blunter. "We were the worst IPO of 2017, I believe." Five weeks after the listing, on June 6, Bloomberg ran a story under a headline that read like a list of charges: "Cars in Vending Machines, a Fading IPO and an Ex-Con Owner." The prospectus had not mentioned that the company's controlling shareholder, the man at the center of the photograph, had pleaded guilty to felony bank fraud in October 1990 for his part in the collapse of Charles Keating's Lincoln Savings and Loan. John Coffee, a corporate-law professor at Columbia, told the reporters that disclosure might not be legally required. "A controlling shareholder having a fraud conviction is of interest to other shareholders," he said. Ernie Garcia III declined to comment. Ernest Garcia II declined to comment.
That last pair of sentences, or something like them, recurs throughout the public record of the Garcias for the next nine years. Both Garcias declined to comment. The son eventually became one of the more available chief executives in American business. He appears on podcasts, at industry roundtables, on CNBC, at luncheons in Phoenix where he is handed carved wooden figurines. The father almost never speaks. The record contains a great deal about what Ernest Garcia II has done (bought, sold, pleaded, financed) and almost nothing about what he has said.
Even the son's name won't hold still. Automotive News profiled him in December 2015 as "Ernie Garcia Jr." A dealer-industry podcast introduced him the same way in December 2024, under the title "The Rise and Fall and Rise Again of Carvana." Wikipedia redirects "Ernie Garcia Jr." to "Ernest Garcia III," which is technically correct, since the original Ernest Garcia was a liquor-store co-owner and onetime mayor of Gallup, New Mexico, and this Ernest is the third. Still, the slippage is telling. People keep demoting him by one generation, folding him back into his father, as if the numeral were a claim still under review.
This is the story of a man who built a company worth tens of billions of dollars, watched it lose 99 percent of its value, and then brought it back to the S&P 500, at $20.3 billion in revenue and $1.41 billion in net income in 2025. Underneath that story is an older and more obvious question. Every profile of him eventually asks it, usually in the third paragraph. How much of what a son builds is his, and how much did he inherit along with the name?
By the Numbers
The Carvana Arc
$15 → $8IPO price in April 2017, and where the stock traded within its first week
$370.10All-time closing high, August 2021
$3.55All-time closing low, late December 2022
98%Decline in Ernie Garcia III's net worth during 2022, per Bloomberg, to $119 million
$3.6B+Carvana stock sold by Ernest Garcia II between October 2020 and August 2021
$20.3BCarvana revenue in 2025, with $1.41B in net income
23,100Carvana employees in 2025
What the Father Built From Wreckage
Gallup sits on old Route 66 in McKinley County, against the Navajo Nation, a railroad-and-trading-post town where commerce has always meant small margins and rough customers. Ernest Garcia II was born there on May 1, 1957. His father, also Ernest, co-owned a liquor store with a man named Frank Colaianni and served for a time as the town's mayor. The son went to the University of Arizona, played golf, dropped out (accounts differ on whether he later finished the degree), and became a stockbroker and then a real estate developer in the Arizona of the 1980s. Money was cheap then, the savings and loans were newly deregulated, and desert land could be financed into something that looked like value.
In October 1990, Garcia, then a Tucson-based developer, pleaded guilty to a felony bank fraud charge. According to the Los Angeles Times, he had "fraudulently obtained a $30-million line of credit in a series of transactions that also helped Lincoln hide its ownership in risky desert Arizona land from regulators." He had acted as a straw borrower, a man whose name went on paper so that someone else's exposure could disappear. Lincoln's collapse became a national scandal, and five United States senators were tainted by their association with Keating. Garcia avoided prison by testifying for prosecutors against Keating. He received three years of probation and was barred from the financial industry for at least three years. He and his firm filed for bankruptcy.
He was thirty-three, and nearly everything a man in his position could lose, he had lost.
In 1991, the year after the plea, he bought a bankrupt rent-a-car franchise called Ugly Duckling for less than $1 million. He merged it with a small finance company of his own and turned it into a business that sold used cars to people whose credit histories would get them turned away at a franchise dealer, and then lent them the money to pay for the cars. Its customers were subprime borrowers. Its product was nominally automobiles but really credit, extended to people nobody else would extend it to, and priced accordingly. In 1996 he took the company public on the Nasdaq under the ticker UGLY. It is hard to read that ticker as anything but a joke the founder was in on. In 1999 he faced six lawsuits alleging he had "abused his position to profit" from a real estate transaction in which he acquired 17 company properties at a 10 percent discount. In 2002, with Ugly Duckling's former chief executive Gregory Sullivan, he took the company private and renamed it DriveTime. Sullivan, a numbers man, would later go on to co-found AFAR.
DriveTime grew into one of the largest used-car retailers in the country, with roughly 150 locations. Its chief executive for years was Raymond Fidel, a former bank president who had pleaded guilty to securities fraud in the same Lincoln affair and had likewise testified against Keating and served three years of probation. Two men convicted in the same thrift collapse, both with their civil rights restored according to DriveTime filings, ran a company whose business model rested on lending to people the system had marked as risks. You can read that as irony or as a kind of fluency. People who have been on the wrong side of a balance sheet tend to understand what a second chance is worth, and how to charge for one.
In time Forbes would put the arc in a headline: "How An Ex-Con Became A Billionaire From Used Cars." An earlier Forbes piece about the lawsuits had been titled "Feathered Nest." By May 2026, the man from Gallup was estimated to be worth $23.9 billion, the richest person in Arizona. He owned an apartment in Trump Tower. His real estate firm, Verde Investments, had bought an 80-acre former mall site in Mesa for a mixed-use development called Fiesta Redefined. The Garcia Family Foundation gave to Arizona schools. Nearly all of that fortune came from a company his son started and on which he held no title.
A Quant on the Subprime Lot
The son grew up in the Phoenix area while his father was rebuilding. That history is easy to recite and hard to picture: a boy of eight or nine whose father had just pleaded guilty to a felony and was spending his probation turning a bankrupt rent-a-car chain into a subprime lender. What the boy absorbed is unrecorded. What the young man did next is documented.
He went to Stanford and graduated in 2005 with a bachelor's degree in Management Science and Engineering. The department is a hybrid of operations research, economics, and decision theory, and it produces people who think of a business as an optimization problem with constraints. In 2025, twenty years after he left, Carvana became the official auto retailer of Stanford Athletics. It is the sort of detail that means everything or nothing.
From 2005 to 2006 he worked as an associate in the Principal Transactions Group at RBS Greenwich Capital, specializing in consumer credit-based investments. These were the years when Wall Street was packaging American consumer debt into securities at a pace and with a confidence that would end badly two years later. Garcia was on the buying side of that machinery, learning how loans become assets and how assets become something you can sell to someone else.
In January 2007 he went home and joined DriveTime. His titles there read like a curriculum:
- Financial strategist, January 2007 to December 2008.
- Managing director of corporate finance, December 2008 to November 2009.
- Vice president and treasurer, and also director of quantitative analytics, November 2009 to January 2013.
In the last role he was responsible for the continued development of DriveTime's consumer credit scoring models and for their use in deal structuring and vehicle price optimization. In plain terms, he built the math that decided how much DriveTime would lend to a given borrower, on what terms, for which car, at what price.
Hold onto that, because nearly everything later written about Carvana skips over it. Carvana is usually described as a technology company, the "Amazon of used cars," with photographs of glass towers, a slick checkout, and a seven-day return window. Its founder learned the business inside a subprime lender, as a quant whose job was to price risk on people with damaged credit. Years later, a short seller's central allegation was not about the glass towers. It was about the loans: how they were underwritten, to whom they were sold, and how the gains from selling them showed up in the income statement. The son's formation and the son's controversy turn out to be the same subject.
Ryan Keeton and Ben Huston, the two men who co-founded Carvana with him in 2012, rarely surface in the coverage except in lists. Huston became chief operating officer, Keeton chief brand officer. Along with Garcia, both were named to Fortune's 40 Under 40 in 2017. Fourteen years after the founding, both were still at the company. That fact is not glamorous, but it matters in a story about a company that nearly died.
Glass Towers by the Freeway
Carvana began as a subsidiary of DriveTime. Its first money came from DriveTime, roughly $100 million of it according to Car and Driver. Even so, the son has said his father was skeptical.
"Probably a different company would swing at [this idea] every couple of years until someone eventually cracked the nut," Garcia told the magazine in 2024. "Until that happened, there would just be failures, and people would probably think, 'Oh, see, it doesn't work.' When the reality was, it was just really hard to build something that was good enough to where consumers said, 'That was good, and I'm going to tell my friends.'"
The idea itself was not original, and he never pretended it was. Buy a used car online. Have it delivered. Keep it seven days and return it if you don't like it, which unwinds the entire transaction, financing included. No salesperson, no haggling, no afternoon lost under fluorescent lights while someone "talks to my manager." Carvana launched in Phoenix in 2013 and called itself the "first complete online auto retailer." It went after an industry that had barely changed in most of a century. At an Economic Club of Phoenix luncheon on November 1, 2017, Garcia traced the problem to the 1930s, when manufacturers decided to let independent dealerships sell their cars. "That's led to 63,000 dealerships, and they can't differentiate the product," he said. Each dealer carried heavy costs for real estate, staff, and advertising, and each had to promote prices it could not sustain. Then he said the line that every reporter in the room wrote down: "There's a reason why culturally if someone calls you a used-car salesman, it's not a compliment."
The son of a subprime used-car magnate said this in public, a few miles from his father's headquarters in Tempe. He didn't appear to notice the irony, or he noticed and didn't care.
"We wanted to exploit the massive opportunity we thought existed," he continued. Then he argued that the opportunity was not really about cars. "Our view is it doesn't have much to do with the product. The customer experience is at the root." He put numbers on it: Carvana saved customers $1,500 per car and four hours of their time. "We're not curing cancer," he said, "but we save customers $1,500 every time they buy a car, which they then can spend on their families. We save them four hours, which they can spend with their families. Those are meaningful things."
At the end of the event, Amy Hillman, dean of Arizona State's W. P. Carey School of Business, gave him a kachina, a carved figure in the Hopi tradition. The club gives one to every speaker. Hillman chose an owl because, she explained, it represents intelligence and wisdom.
The company's real cleverness was the vending machine. The first version appeared in 2013. In 2015, in Nashville, Carvana opened the version that made it famous: a multistory glass tower, eight stories in some cities, holding around 30 cars, fully automated and coin-operated. A customer who picked up a car instead of taking home delivery received an oversized Carvana token and dropped it into a slot. Machinery hummed, a platform slid, and the car descended to a delivery bay like a can of soda. By 2017 there were towers in seven cities. By April 2020 there were 24, and later more than 30. Carvana patented them.
People called them a gimmick, and they were one, in the way a good gimmick is also an economic argument. "The vending machine builds advertising into the model," Garcia told the Phoenix audience. "It sits by the freeway. It gets people talking. It's interesting. It represents what we do." It was also cheaper than delivering to the customer's door. The building was a billboard, a warehouse, and a delivery depot at once, and its glass walls displayed the inventory to every commuter on the interstate.
The Miami Herald, reporting on the tower off I-95 in Hollywood, Florida, in 2023, called it "a new gimmicky landmark." It did not occur to the headline writer that being called a landmark was the point.
None of this made Carvana easy to finance. "What's funny in the story of 2013 through 2016," Garcia said, "was that [investors believed] we had no chance. We started out in Phoenix, and that was the wrong place to be building a startup. And we are a capital-intensive business at a time when the only businesses getting funded were marketplaces. We got to a spot where, basically, private capital dried up for us." Venture investors in those years wanted asset-light platforms that matched buyers and sellers and took a cut without touching inventory. Carvana wanted to own the cars, recondition them, finance them, and deliver them on its own trucks. A rival called Beepi raised $150 million in venture funding and then wound down in early 2017. The money that kept Carvana alive was family money, which is a different thing from venture money. It comes with fewer term sheets and more history.
The Worst IPO of 2017
So they went public, early and out of necessity. "With more money needed for growth," Car and Driver noted, "in 2017, Garcia III (the CEO) and Garcia II (the privately held company's biggest shareholder) took the online retailer public." The offering raised $225 million. The structure Wall Street was asked to accept had three main features.
- Control. Super-voting Class B shares meant the family could sell a great deal of economic ownership without giving up command.
- A tax receivable agreement. Most of any tax benefit generated in the IPO, estimated at the time at $1 billion, would be passed to early investors. The largest of those was Ernest Garcia II.
- Related-party ties. The Bloomberg reporters called them a "spaghetti pile." Carvana did business with DriveTime, with DriveTime's loan servicer, and with other entities the father controlled.
The prospectus warned investors about this in the flattened language securities lawyers use. "The interests of the Garcia parties may not in all cases be aligned with your interests," it said. The sentence would be quoted back at the company for years.
By the time Bloomberg's June 6 story ran, Carvana's shares had fallen 36 percent from the IPO price and closed that day at $9.53. The company was operating in 27 cities, burning cash, and had doubled both its revenue and its net loss in the first quarter. That pattern of growth running ahead of losses would last until it nearly killed the company.
On August 8, 2017, Garcia signed the first of his quarterly Sarbanes-Oxley certifications as chief executive of a public company. Every public-company CEO signs the same boilerplate: I, Ernie Garcia, III, certify that... based on my knowledge, this report does not contain any untrue statement of a material fact. He signed it "/s/ Ernest C. Garcia, III," with a comma before the numeral. He would sign it again every quarter for the rest of the decade.
The company grew fast. In the third quarter of 2018, retail units sold rose 116 percent year over year, total revenue rose 137 percent, and gross profit rose 181 percent. That made 19 consecutive quarters of triple-digit unit and revenue growth, with a 20th on the way. "To illustrate the power of that compounding growth," the shareholder letter said, "we sold more vehicles in Q3 than in all of 2015 and 2016 combined." The company bought a rival startup, Calypso, in 2017. In April 2018 it paid $22 million for Car360, a Mark Cuban-backed company whose smartphone software used 3D computer vision and augmented reality to photograph cars. Cuban later called the deal Shark Tank's "No. 1 success story." Around 2019, according to a Wells Fargo senior manager quoted years later, the bank called off an agreement in principle to buy Carvana's loans. "As we dug into it," the manager said, "the more we learned, the less we liked about it."
Hardly anyone outside the bank heard about that at the time. "By the time we got to 2019," Garcia recalled, "people sort of believed in the business. It made sense. People thought maybe we were onto something."
In September 2018 the company sold its 100,000th car, and its chief executive did something chief executives rarely do: he gave away his own stock.
The mechanics are worth spelling out, because in this case the mechanics were the gesture. Garcia committed roughly $35 million of his personal Carvana shares to the company's employees. Each employee would receive 165 restricted stock units on reaching their one-year anniversary. At the closing price of $68.28 on Monday, September 10, 2018, each grant was worth just over $11,000. For a reconditioning technician or a flatbed delivery driver, that was a meaningful sum. Because the company formally issued the grants, they would show up as compensation expense: about $11 million in the third quarter and $3 million in the fourth, at that price. For every share the company granted, Garcia would contribute one of his own shares back to the company to be retired, so the number of shares outstanding would not change. He was paying the bill himself, one share at a time, and filing a Form 4 with the SEC each time he did it.
The shareholder letter included his explanation, which reads like it was written by a person rather than a communications department:
— Ernie Garcia III, Carvana Q3 2018 shareholder letterIt can get easy to zoom in too far, too fast and lose sight of what really matters. In the long run, I believe what matters most is that we have a concept that makes our customers' lives a little bit better, that we attract and retain great people, and that those people care about what they are doing. We have great people. And they care.
"We wouldn't be where we are today without them," he added, "and I certainly wouldn't be where I am." The company tied the gift to one of its stated values, "We're All in this Together." Values statements are usually wallpaper. This one had been backed by $35 million in stock.
It's tempting to read the gift through the family history. A son whose father's fortune rested partly on arrangements other shareholders considered one-sided, such as tax agreements, discounted properties, and related-party deals, went out of his way to transfer value downward, at his own cost and with no dilution to outside investors. Maybe it was a deliberate contrast. Maybe it was simply generosity, of the kind easily afforded by a man whose stake was worth billions on paper. Both can be true. The record shows only the gift and the accounting.
The record also shows how he talks about the people he hires. "Great people don't want to come to work every day if they're not inspired and achieving something in a positive way," he said in Phoenix. Years later, in an interview, he gave a sharper definition of culture than most management books manage: "What matters is when you're sitting in a room of six people and you're doing things. What is normal in that room — that's what your culture is." He has also described real mentoring in terms that sound more like an apprenticeship than a seminar: "Real mentoring isn't about sitting in a room and telling people the rules. It's about being there solving a problem and having a person watch the problem be solved."
He can be frank about himself, too. "I'm excitable," he has said, "and I want to work on the next cool thing. My instinct is to jump to the next thing too fast. Prioritization and follow-through are very, very important, and I think that's not my relative strength." Few founders say that about themselves in public. In hindsight, it also explains a good deal about 2021.
Cheyenne Trimels shows what that culture looked like from below. She graduated from Hillsdale College in 2017 and taught middle and high school history. In early 2020 she left teaching for a job in Carvana's government affairs group, where she eventually led the company's legislative work. "Buying a car is a big deal," she told her alma mater's blog, "and making that experience better is something that does actually help people." She is a former history teacher whose job became explaining a new company to state lawmakers. She took that job just before the states started suspending Carvana's licenses.
Selling Into the Bloom
When the pandemic arrived in March 2020, the first weeks were frightening. "When Covid first hit, that was a scary moment for sure," Garcia said later. "But luckily that only lasted a couple of months. And then it turned into a positive." On April 15, 2020, he went on Jim Cramer's Mad Money. Other auto retailers had reported sales declines of 50 to 85 percent in the second half of March. Carvana, he said, had "held up significantly better than that." The company had switched to touchless delivery and was offering 90-day payment deferrals. He predicted, accurately, that the federal stimulus checks then going out ($1,200 per individual, $2,400 per couple) might lift demand. "I do think in general when people get big sums of money from the government, whether it's tax season or now with this stimulus, that does generally drive some large-ticket purchases," he said. Carvana's shares closed that day at $71.34, giving the company a market value of $11.8 billion.
Around then, while the shares were depressed, Ernest Garcia II bought Carvana stock in a private offering at a price below market. The shares doubled within two months. Unhappy shareholders sued.
For the rest of 2020, every trend broke Carvana's way. People stopped wanting to sit in dealerships. New-car supply ran short and pushed up used-car prices. Interest rates were close to zero and stimulus money was everywhere. Carvana sold 244,111 vehicles that year on $5.587 billion in revenue, becoming the second-largest online used-car retailer in the country. In 2021 it made the Fortune 500, one of the youngest companies ever to do so. On August 10, 2021, the stock closed at $370.10, an all-time high, and the company was valued at nearly $60 billion. Father and son together were worth more than $32 billion.
The father had already started selling. In late October 2020, when the shares were near $200 (they had been about $90 before the pandemic), Ernest Garcia II began selling Carvana stock. His first sale netted roughly $388 million. Over the next ten months, according to Bloomberg's reading of his SEC filings, he sold "almost every day." By September 2021, the Wall Street Journal reported, he had sold more than $3.6 billion, 16 percent of his holdings. Only the founders of Amazon and Facebook and the heirs of Sam Walton had made more selling stock in their own companies over that period, according to the analytics firm InsiderScore. Those companies had earned more than $106 billion in combined profits over the previous six quarters. Carvana had lost roughly half a billion dollars, with one profitable quarter of $45 million that spring.
His last sale came on August 23, 2021. That was about two weeks after the peak.
— Daniel Taylor, Wharton School Forensic Analytics Lab, to the Wall Street Journal, 2021The existing structure has allowed them to run this $60 billion public company as if it's a family firm and for the family's benefit.
"It's amazing," Taylor added. Carvana's spokeswoman said the company had robust governance, followed all rules and regulations, and had returned more than 2,000 percent to investors since the IPO. Both Garcias declined to comment. The tax receivable agreement meant Carvana could owe Garcia II the bulk of $1.1 billion. In December 2021, the Journal reported that Carvana had bought thousands of cars from DriveTime to meet pandemic demand, and that it had not disclosed the younger Garcia's significant stake in DriveTime and in other companies that supplied services to Carvana. A Carvana spokesperson said working with affiliated companies gave the firm "a unique advantage" and allowed it to grow faster.
That is the uncomfortable center of this story. Nothing in the public record shows the son selling. In 2024, when Carvana's shares had recovered, Fortune reported that his father had sold $12 million worth that year and that he had sold less than $1 million. In 2022, at the bottom, he put his own money in. The securities class action that followed nevertheless names him, and its theory relies on the family relationship. Filed in the District of Arizona and led by the United Association National Pension Fund and the Saskatchewan Healthcare Employees' Pension Plan, the suit alleges that false and misleading financial statements issued between May 2020 and July 2022 inflated the stock price while the father sold. In March 2026, as a class-certification deadline approached, the plaintiffs asked for communications about what the son knew of his father's sales and whether he benefited from them. They argued he stood to inherit the proceeds through a family trust. Carvana answered that the sales were made individually and that extensive discovery had turned up no evidence of misconduct. U.S. Magistrate Judge John Z. Boyle said he would review the trust documents himself before ruling.
So a federal magistrate in Phoenix proposed to read the instrument by which one Ernest Garcia's wealth might pass to the next, in order to decide whether the son's conduct as a public-company CEO could be separated from his father's conduct as a seller. The question every profile asks had become a discovery dispute.
Massively Overextended
"Late '21, a lot of things happened," Garcia said later. "We were massively overextended."
What happened was fairly simple. Supply chains seized up and car prices climbed. The Federal Reserve began raising rates, which made financing a used car more expensive, which hurt demand, which left Carvana with expensive inventory bought at the top. Carvana had built capacity for a growth curve that ended abruptly. The company had chosen, by its founder's own account of his temperament, to jump to the next thing fast.
Then it made the move nobody has let it forget. In February 2022, Carvana agreed to buy ADESA's U.S. physical auction business, the country's second-largest wholesale auto auction chain, from KAR Auction Services for $2.2 billion. The deal brought 56 auction sites, acres of land near population centers, and reconditioning capacity Carvana would otherwise have had to build. It also added about $3.2 billion of debt at the moment investors stopped rewarding growth and started demanding cash flow. "They pursued an aggressive growth strategy, both organic — opening three new inspection and reconditioning centers — and inorganic, buying KAR's 56 auction sites, right as used car end demand tailed off aggressively given rising interest rates," Chris Pierce, an analyst at Needham & Company, told Modern Retail. By January 2023, Carvana's total debt exceeded $7.4 billion.
The collapse unfolded in a series of hard dates.
On April 20, 2022, Carvana reported a first-quarter loss of $506 million and described a "uniquely difficult environment." It announced a $1 billion stock offering and another $1 billion in preferred stock. Garcia was one of two investors who indicated interest in buying as much as $432 million of the common shares. The son bought as his father had sold. By then he had lost 60 percent of his net worth since the start of the year, about $4.1 billion, the steepest decline of any American billionaire Bloomberg tracked, ahead of even Netflix's Reed Hastings.
On May 5, the stock fell 18 percent in one of the worst trading days in two years and stood 87 percent below its August peak. The family's combined fortune had lost almost 80 percent in nine months. CAS Investment Partners, the hedge fund run by Clifford Sosin, had made a large bet on Carvana, and it went badly. Tiger Global, D1 Capital, Whale Rock, Marshall Wace, and Sculptor had all held big stakes at year-end. Plenty of sophisticated investors had trusted the company.
On May 10, Carvana laid off 12 percent of its workforce, about 2,500 people, mostly in Phoenix, by Zoom. Forbes called the firing "chaotic." On the same day, Illinois suspended Carvana's dealer license, effectively banning it from the state. The Secretary of State said the company had failed to deliver titles to buyers within the 20 days state law required, and in some cases had issued out-of-state temporary registrations to Illinois residents. The state lifted the suspension on May 26 under strict conditions, then reimposed it on July 18 after finding the illegal conduct had continued. Paul Breaux, Carvana's vice president and general counsel, was charged with multiple misdemeanor and petty offenses. Carvana won a temporary injunction on July 29. North Carolina had already temporarily barred the Raleigh vending machine in August 2021 over inspections and titles. A class action in Pennsylvania alleged unfair and deceptive practices in transferring ownership, which Carvana denied. In June 2022, Barron's ran a story headlined "Carvana Sought to Disrupt Auto Sales. It Delivered Undriveable Cars." On October 7, Michigan suspended the license of Carvana's dealership in Novi, citing "imminent harm to the public." The state had already put the company on probation in May 2021 and extended that probation in February 2022 after further violations. The Novi location kept operating the next week. Carvana called the regulations "outdated" and the suspension "baseless and reckless."
The irony is close to unbearable. The company's pitch was the removal of friction, and it failed at the most bureaucratic and least glamorous part of selling a car: paperwork. A customer could drop a giant coin into a glass tower and watch a car descend, then wait weeks for a title and be unable to register the vehicle. The front end was spectacular. The back end did not keep up.
On August 17, 2022, Garcia appeared at the Automotive Remarketing Alliance's Summer Roundtable in Nashville, the city of the first coin-operated tower, in front of an audience that included auction owners who were now, through ADESA, his colleagues. "We guessed wrongly what this retail year would be like and found ourselves out of balance and had to make changes," he said. "I'm a massive optimist, and I think to make progress in anything, you have to get knocked down many times, take the pressure, turn it around, and make it something positive. We plan to get up and kick back. Give us time."
He didn't get much. On November 4, after third-quarter results showed retail units down 8 percent to 102,570, the stock fell about 40 percent in a day. Another 1,500 employees were laid off that month. On December 6, the website 24/7 Wall St. named him a finalist for Worst CEO of 2022, alongside Mark Zuckerberg. Its write-up noted that Bank of America had said the stock could go to zero and concluded: "It would be hard to find a CEO who wrecked a company so quickly." The next day, December 7, Carvana's largest creditors signed an agreement to negotiate together in any debt restructuring. Lawyers call that a cooperation agreement. Everyone else read it as a sign that bankruptcy was close. The stock fell 42 percent that day, and Garcia's fortune fell with it by $87 million. By Bloomberg's count he had lost $6.7 billion that year and was down to $119 million, a 98 percent decline. Among everyone the index tracked, only Sam Bankman-Fried had lost a larger share of his wealth in 2022.
Late that month the stock closed at $3.55. On January 17, 2023, Carvana adopted a shareholder rights plan, a poison pill, meant to protect its net operating losses from an ownership change that could jeopardize them. Put plainly, the company was trying to preserve the tax value of its own losses. That month the Journal reported Carvana was quietly letting employees go, cutting hours, and leaving open positions unfilled.
The losses were not the founder's alone. Among the people hurt were the employees who received 165 restricted stock units as a gift in 2018, worth more than $11,000 then and worth, at $3.55 a share, $585.75.
The Arithmetic of Not Dying
How Carvana survived is less dramatic than how it nearly died, and more instructive. It came down to arithmetic and software, done for eighteen months by people who had stopped being able to afford anything else.
The company restructured its debt in 2023, swapping near-term obligations for longer-dated ones. In late August 2023, father and son increased their equity stakes, and the shares rose more than 10 percent the next trading day. In February 2024, CNBC summarized the work. Headcount had been cut by more than 4,000. About $1.1 billion in annualized expenses had been removed from operations. Carvana had built proprietary software called Carli to run vehicle reconditioning and other formerly manual tasks end to end, built logistics tools for mapping, route optimization, and driver scheduling, and started using generative AI for certain tasks. The pursuit of growth gave way to the pursuit of gross profit per unit, the dollars kept on each car after reconditioning, transport, and sale. The ADESA sites, a burden in 2022, became the company's reconditioning capacity. Garcia later described how the turnaround felt from inside, emphasizing what he called "ruthless prioritization," a phrase that sounds like a correction applied to someone who knows his instinct is to jump to the next thing too fast.
The market was skeptical. On June 8, 2023, Carvana said it expected about $50 million in adjusted EBITDA for the quarter, and the stock jumped 56 percent, topping $25. By the close the next day it was back to $19.07. TechCrunch noted that profitability was improving even as revenue fell: analysts expected $2.57 billion for the second quarter, against $3.88 billion a year earlier. Long-term debt exceeded $6.5 billion. Operating cash flow in the first quarter had been negative $66 million.
He kept the same tone throughout. When Patrick O'Shaughnessy had him back on Invest Like the Best after the worst had passed (they had first talked in 2021, at the peak), O'Shaughnessy said Garcia had "the exact same demeanor and crystal-clear vision for the business" as before. He added, "I don't know many leaders who could survive this." In those conversations Garcia kept crediting the team, and kept describing a way of handling crisis that reads like a game plan. "We try to tell that story," he said. "We try to set that up ahead of time. We try to point to other companies that we all have a ton of respect for." Before anything went wrong, he said, he had told his people that companies attempting something hard almost always nearly die along the way, so that when Carvana's turn came it would feel expected rather than catastrophic. He described the early hires as people with "a sports mentality." He argued that the senior team stayed together because its members were not primarily motivated by status. When a company falls apart, status is usually the first thing people try to protect.
He has never claimed equanimity he didn't feel, though. At the Automotive Remarketing Alliance roundtable two years after Nashville, on August 21, 2024, at the PGA Omni resort outside Dallas, he went on stage with John Combs, a vice president at ADESA, and spoke more openly than executives usually do:
— Ernie Garcia III, ARA Summer Roundtable, August 21, 2024When you hear the stories from a distance, it always sounds sort of heroic. When you're in those moments, it's a real bummer because you wake up every day with a million things to do and everyone questioning. You see people around you, I hate to say, who kind of gleefully enjoy your difficulty, and that's tough to see. You wish it weren't true about the world, but it is.
The stock closed at $159.78 two days later. On May 2, 2024, after the company reported its strongest quarter ever and its first year-over-year sales growth in six quarters, shares had jumped 34 percent in a day. Measured from the December 2022 low, the stock had risen more than 3,000 percent. Bloomberg calculated that father and son had added more than $11 billion to their combined net worth. The father's fortune went from a 2022 low of $3.1 billion to $10.9 billion, the son's to $3.8 billion. "2022 and 2023 were a tough run for us," Garcia said on Bloomberg TV that day. "When we went through that period, the team came together, we responded incredibly well and I think this quarter is undoubtedly the best quarter we've had in our history."
He was 42. Fortune's first version of the story described the father as a co-founder. A correction followed: "Corrects to remove reference to Ernie Garcia II being a co-founder." Once again, a caption had folded the son into the father, and it had to be fixed.
We Don't Talk About DriveTime
On January 2, 2025, Hindenburg Research published a report on Carvana under the title "A Father-Son Accounting Grift For The Ages."
Hindenburg said the report drew on four months of work and 49 interviews with former employees, competitors, industry experts, and related parties. It called the turnaround "a mirage." It noted that Carvana's stock had risen 284 percent in 2024, to a market value of about $44 billion, and that for every dollar of net income the company had reported, it had added $139 in market value. It said almost 26 percent of Carvana's gross profit came from selling customer auto loans to third parties, largely subprime and deep subprime, and that gains on those sales amounted to 2.2 times net income over the previous nine months. It said Ally Financial had bought $3.6 billion of Carvana loans in 2023, about 60 percent of originations, and had amended the arrangement five times in two years. It alleged $800 million in loan sales in 2024 to "a suspected undisclosed related party," which it later identified as likely a trust affiliated with Cerberus Capital Management. Cerberus's chairman of global investments is Dan Quayle, the former vice president, who also sits on Carvana's board. It quoted former employees describing underwriting standards as nearly nonexistent. It noted that the father had sold another $1.4 billion of stock after the 2023 rally, almost 10 million shares in 2024.
It also quoted a former director describing the relationship between Carvana and the father's company in terms borrowed from a movie: "There's certain things we don't talk about, and we don't talk about DriveTime."
Carvana called the report "intentionally misleading and inaccurate." The shares fell about 6 percent the next day and closed below $200 for the first time since October. The son lost $115 million on paper and was still worth more than $6 billion. The father lost $173 million.
What followed was, in market terms, anticlimactic. The stock did not collapse. By July 2025 it had reached $334, about $39 billion in market value. In 2025, Carvana reported $20.3 billion in revenue, $1.88 billion in operating income, and $1.41 billion in net income, with 23,100 employees, and it joined the S&P 500. It began selling new cars, a move the Wall Street Journal reported in May 2026 "has dealers rattled." By early 2026 the son's fortune was roughly $10 billion. His father kept selling through planned transactions, filing to sell more than $500 million in the first half of 2025, converting Class B voting shares to Class A and liquidating them.
The lawsuit continued. Motions to dismiss were denied and so was the motion for reconsideration, and the case moved into discovery. By March 2026 the parties were also fighting over access to records from SEC investigations in 2019 and 2025. There was the question of the trust.
An investment writer named Ryan Reeves, after a long analysis of the business in May 2025, set aside the arithmetic to address what he called the charge most often thrown at the son. He quoted Jeremiah 31:29–30: The fathers have eaten sour grapes, and the children's teeth are set on edge. The prophet's point is that this saying will no longer hold, and that each person will answer for his own sin. "Ernie Garcia III is not his father," Reeves wrote. "He's his own person. Move on."
The market has more or less agreed. The courts have not yet decided. The structure of the company argues in both directions. According to Reeves's count, the son owns about 14 percent of the shares and the father about 20 percent, and together, through the super-voting shares, they control roughly 85 percent of the votes. That is less than the 97 percent of 2017, but it is still a family firm in every sense that matters for control.
There are smaller, more personal details too. In the summer of 2024, an LLC tied to a Tempe-based family trust overseen by the younger Garcia paid the Spearman family nearly $6.3 million for the Racquet Club of Irvine, an old club with tennis and pickleball courts that local residents had long fought to keep from developers. Carvana has been the title sponsor of the Professional Pickleball Association Tour since 2022. It sponsors Jimmie Johnson's race cars and the youth programs of Phoenix Rising FC. Nothing in the record says whether the trust that bought the racquet club is the one Judge Boyle asked to read. The word simply keeps coming up.
At some point in those years, Arizona Business Leadership invited him to speak, and its board president presented him with the group's traditional gift. It was a kachina, as in 2017. The organization's post did not say which figure it was.
The certifications kept coming every quarter, as the law requires. The language has not changed since August 8, 2017: Based on my knowledge, this report does not contain any untrue statement of a material fact. At the bottom, typed rather than handwritten, as it is on every one, is the name his grandfather took to the mayor's office in Gallup and his father put on a straw loan in the Arizona desert:
/s/ Ernest C. Garcia, III
Part IIThe Playbook
Ernie Garcia III's career is unusually useful to study because it includes both extremes. He built a company in an industry that had resisted change for most of a century, nearly destroyed it through overextension, and then rebuilt it into one of the most profitable automotive retailers in the country. He did all of this under a governance arrangement that critics call the defining risk of the company and defenders call the reason it survived. The principles below come from that record. Some are worth copying. A few work better as warnings.
The principles
- 1.Sell the experience; the product is a commodity.
- 2.Build the advertising into the infrastructure.
- 3.Know where your edge came from, because someone else will look.
- 4.The back office is the product.
- 5.Tell the crisis story before the crisis.
- 6.Pay your people out of your own pocket.
- 7.Name your weakness, then staff against it.
- 8.Separate a broken plan from a broken thesis.
- 9.Buy when you are being written off.
- 10.Turn the cost cuts into software.
- 11.Culture is what's normal in a room of six.
- 12.Control is an asset with a carrying cost.
C
The Carvana Arc
Key moments in the company's rise, collapse, and recovery.
1990
Ernest Garcia II pleads guilty to felony bank fraud in the Lincoln Savings collapse.
1991
Garcia II buys bankrupt Ugly Duckling for under $1 million; it later becomes DriveTime.
2007
Garcia III joins DriveTime, eventually directing its consumer credit-scoring models.
2012
Carvana co-founded as a DriveTime subsidiary with Ryan Keeton and Ben Huston.
2015
First fully automated, coin-operated vending machine opens in Nashville.
2017
IPO at $15; trades at $8 within a week. Founder calls it the worst IPO of the year.
2018
Garcia III gives about $35 million of personal stock to employees at the 100,000th car.
2021
Stock peaks at $370.10 in August; Garcia II's selling, begun in October 2020, ends on August 23.
2022
ADESA acquired for $2.2 billion; layoffs; state license suspensions; stock closes at $3.55.
2023
Debt restructured; more than $1 billion in annualized costs removed; Garcias add to their stakes.
2025
Hindenburg short report; Carvana posts $1.41 billion net income and joins the S&P 500.
Principle 1
Sell the experience; the product is a commodity
Garcia's most important strategic insight came in one sentence: "Our view is it doesn't have much to do with the product. The customer experience is at the root." A used 2016 Camry is the same Camry whether it sits on a dealer lot in Mesa or a reconditioning line in Tempe. Garcia's own diagnosis of the industry was that 63,000 dealerships "can't differentiate the product." When the product can't be differentiated, the experience is all that's left to compete on.
This is less obvious than it sounds. Most disruptors in physical goods focus on price and try to be cheaper. Garcia did quote a savings figure, $1,500 per car, but he gave equal weight to four saved hours and treated the seven-day return window as the core of the offering. The return policy substitutes for the test drive and moves the risk of a bad purchase from buyer to seller. It is expensive to provide, which is exactly why incumbents protecting their margins couldn't match it.
Tactic: Identify the part of your industry's transaction customers hate most, and make removing it the thing you charge for, even if it costs you more to deliver.
Principle 2
Build the advertising into the infrastructure
The vending machine is usually treated as a stunt. Garcia described it as a cost structure: "The vending machine builds advertising into the model. It sits by the freeway. It gets people talking. It's interesting. It represents what we do." One glass building handled storage, customer pickup, last-mile logistics, and brand awareness, and passersby got to watch the inventory through its walls.
The general lesson is that operating spending and marketing spending don't have to be separate line items. A facility that is also a landmark spreads its cost over several jobs. The press called the towers "gimmicky landmarks," and "landmark" was the operative word. Customers navigated by them.
Tactic: Before you approve a marketing budget, ask whether some physical or operational asset you already need could do that job just by being visible.
Principle 3
Know where your edge came from, because someone else will look
Carvana's early survival depended on things a typical startup doesn't have. It had about $100 million in family capital when venture money had dried up. It had DriveTime's infrastructure, its inventory during the pandemic shortage, and its loan servicer. It had a founder who had spent six years building subprime credit models. Each of these was a real competitive advantage, and each later became an accusation. Bloomberg found the father's conviction missing from the prospectus. The Wall Street Journal found the son's undisclosed stakes in suppliers. Hindenburg found what it called "a father-son grift." The company's own spokesperson called the related-party ties "a unique advantage," and they were. They were also the company's biggest reputational liability.
None of this means a founder should turn down inherited advantages. Few rational founders would. It means understanding that every unconventional advantage will eventually be judged by people who don't share your view of it, and disclosing on your own terms is better than having it surfaced by a short seller. "There's certain things we don't talk about" is a phrase for a movie, not for corporate governance.
Tactic: List every advantage your company has that a competitor couldn't buy, and for each, write the hostile headline about it before a reporter does. Then decide what to disclose.
Principle 4
The back office is the product
In 2022, the customer-experience company kept running into regulators over titles and registrations. Illinois suspended its license twice. Michigan cited "imminent harm." North Carolina shut down a vending machine. A Pennsylvania class action challenged its ownership transfers. Barron's reported cars delivered in undriveable condition. Whatever the merits of each case (Carvana disputed several and called some regulations "outdated"), the pattern made one thing clear. The front of the funnel had been designed for delight, while the back depended on 50 different state bureaucracies that didn't care about the user interface.
The company's recovery was largely a back-office recovery: Carli for reconditioning, routing and scheduling tools for logistics, and automation of manual steps. Its competitive advantage moved from the glass tower to the systems behind it.
Tactic: Follow your product all the way to the last government form or warranty claim, and invest in that step as heavily as you invest in the checkout page.
Principle 5
Tell the crisis story before the crisis
Garcia's most practical leadership habit is preparing people ahead of time. "We try to tell that story. We try to set that up ahead of time. We try to point to other companies that we all have a ton of respect for." Long before 2022, he had been telling his team that companies taking big swings almost always come close to dying. "I think most companies that take a big swing generally have multiple near-death experiences," he said. When the stock fell 99 percent, Carvana's employees were watching something they had been told to expect, even if it was worse than they had imagined.
This is different from optimism. Garcia calls himself "a massive optimist," but what protected morale was not a cheerful forecast. It was a shared understanding, built in advance, that adversity was part of the plan rather than proof the plan had failed. "I've made peace with the fact that people are going to question you if you're ever doing anything that matters," he has said. That kind of acceptance has to be in place before it's needed.
Tactic: During your best quarter, tell your team the story of a company you admire that nearly died and recovered, and say plainly that it will probably happen to you too.
Principle 6
Pay your people out of your own pocket
The 100k Milestone Gift worked because of its structure, not just its size. Garcia gave employees about $35 million of his own stock and offset every company grant with a share of his own, so outside shareholders weren't diluted. The company recorded the expense honestly, and the founder covered the cost. The message was clear because the money was his.
The gesture created goodwill that the company drew on four years later, when those same grants had lost most of their value. Notably, the employee-facing value it reinforced was "We're All in this Together," and most of the senior team stayed through the collapse.
Tactic: When you reward employees, consider paying from your own equity rather than the company's, and make the mechanics visible so no one mistakes it for a routine compensation decision.
Principle 7
Name your weakness, then staff against it
"I'm excitable, and I want to work on the next cool thing. My instinct is to jump to the next thing too fast. Prioritization and follow-through are very, very important, and I think that's not my relative strength." Few founders describe themselves this precisely, and the description fits 2021 almost exactly. That year saw massive capacity expansion, the ADESA acquisition, and growth assumptions extended indefinitely. The recovery was in large part a deliberate correction of the founder's natural tendencies: "ruthless prioritization" adopted by an executive who knows prioritization isn't his strength.
Self-awareness alone accomplishes nothing. Its value comes from the structure built around it. Garcia kept co-founders and senior operators around him for more than a decade, and the turnaround was handled by people whose skills complemented his.
Tactic: Write down the one trait that most often leads you into mistakes, and put someone in a position to veto decisions where that trait is likely to drive the outcome.
Principle 8
Separate a broken plan from a broken thesis
"We guessed wrongly what this retail year would be like and found ourselves out of balance," Garcia told the Nashville audience in August 2022. Notice what that statement concedes and what it doesn't. It admits a bad forecast and too much capacity. It doesn't admit that consumers don't want to buy cars online. "Late '21... we were massively overextended" is a confession about execution, not about the idea.
This distinction determined the outcome. Carvana cut 4,000 jobs and $1.1 billion in annual costs, but it kept the ADESA footprint, the seven-day returns, and the vertical integration. Garcia has said that taking on one extra near-death experience is better than sacrificing "the things you deeply believe." The danger, of course, is that every failing company says its thesis is sound. The only test is whether the core unit economics improve once the excess is removed. At Carvana they did.
G
Plan vs. Thesis: What Carvana Changed and What It Kept
The 2022–2024 recovery cut execution while preserving the core bet.
| Element | Decision | Category |
|---|---|---|
| Headcount and growth-era capacity | Cut by more than 4,000 people; $1.1B annualized expenses removed | Plan |
| Advertising and inventory volume | Scaled back sharply in 2022–2023 | Plan |
| Online purchase and seven-day return | Kept intact | Thesis |
| ADESA physical footprint | Kept; repurposed for reconditioning | Thesis |
| Manual reconditioning workflows | Replaced with Carli software and logistics tooling | Rebuilt |
| Growth as the primary metric | Replaced by gross profit per unit and efficiency | Reframed |
Tactic: In a downturn, sort every line of spending into two lists, assumptions about the market and beliefs about the customer, and cut the first list hard before touching the second.
Principle 9
Buy when you are being written off
In April 2022, with his net worth down 60 percent in four months, Garcia indicated he would buy up to $432 million in Carvana's emergency stock offering. In August 2023, when the stock was still far below its peak, he and his father increased their stakes again, and the market responded the next trading day. Whatever else is true about the family's dealings, this is the founder's clearest signal: he bought when the most informed person in the building had every incentive to stay out.
The contrast with his father's selling in 2020–21, near the top, is the most important fact in the governance debate, and it cuts both ways. The son's buying supports his claim of conviction. The father's selling supports the plaintiffs' claim of opportunism. Investors can see both.
Tactic: If you are asking others to believe in your company during a crisis, put a meaningful share of your own liquid net worth into it at the moment it's most painful to do so.
Principle 10
Turn the cost cuts into software
Most turnarounds cut costs and stop. Carvana cut costs by building tools. The $1.1 billion in annualized savings came alongside Carli for end-to-end reconditioning, mapping and route optimization for logistics, driver scheduling systems, and generative AI for certain tasks. The result was a business whose cost per car kept falling as volume came back, because the savings were written into code instead of depending on fewer people working harder.
That is why the comeback compounded. When growth returned in 2024, the company had more operating leverage than before the crash. In 2025, revenue reached $20.3 billion and operating income $1.88 billion.
Tactic: For every role you cut in a downturn, ask what software would let the remaining team handle the same volume, and build it before demand returns.
Principle 11
Culture is what's normal in a room of six
"What matters is when you're sitting in a room of six people and you're doing things. What is normal in that room — that's what your culture is." The definition is useful because you can check it. Culture is not the values poster or the all-hands meeting. It is the unremarkable behavior in a working session with no executive watching. Garcia has suggested a related test for job candidates: watch how junior employees speak in front of senior ones. Confident juniors indicate a healthy organization.
His view of mentoring follows from this. It means solving a problem while someone watches. He also offers an explanation for the senior team's durability: its members aren't primarily motivated by status. When the stock goes to $3.55, status disappears overnight, and people who were there for it leave.
Tactic: Sit in on a routine meeting two levels below you, say nothing, and treat whatever behavior goes unchallenged as your actual culture.
Principle 12
Control is an asset with a carrying cost
The family held 97 percent of voting power at the IPO and roughly 85 percent by 2025. Control let Carvana survive what might have ended a company with activist shareholders or a fractured board. No one could force a sale at the bottom or replace the CEO in December 2022, when 24/7 Wall St. was asking whether anyone had ever wrecked a company faster. Investing City's Ryan Reeves made the founder's case: aside from bankruptcy, a founder's worst nightmare is losing the company.
Control also has ongoing costs. Those costs included a Wharton professor saying the structure let the family run a $60 billion public company "as if it's a family firm and for the family's benefit." They included a tax receivable agreement that could pay the father most of $1.1 billion, a short report built around the word "father," and a federal magistrate reviewing trust documents. Control doesn't remove scrutiny. It concentrates it, permanently, on the people who hold it.
Tactic: If you keep voting control, expect to pay for it with more disclosure than the rules require, and assume every related-party transaction will eventually be read by someone who wants to prove you wrong.
Free playbook
Get The Business Model Playbook
58 business models, one visual page each: how the money flows, the metrics that matter, and who runs it. Free when you join the Faster Than Normal email.
Free. No spam. Unsubscribe anytime.
Part IIIQuotes and Maxims
In their words
— Ernie Garcia III, Economic Club of Phoenix, November 1, 2017There's a reason why culturally if someone calls you a used-car salesman, it's not a compliment.
— Ernie Garcia III, to Car and Driver, 2024Probably a different company would swing at [this idea] every couple of years until someone eventually cracked the nut. Until that happened, there would just be failures, and people would probably think, 'Oh, see, it doesn't work.'
— Ernie Garcia III, interviewI think entrepreneurs are essentially stubborn egomaniacs. You have to have a big enough ego to believe you can change things. And you have to be stubborn enough to keep fighting for it when it gets hard.
— Ernie Garcia III, ARA Summer Roundtable, August 17, 2022I'm a massive optimist, and I think to make progress in anything, you have to get knocked down many times, take the pressure, turn it around, and make it something positive. We plan to get up and kick back. Give us time.
— Ernie Garcia III, Economic Club of Phoenix, November 1, 2017The vending machine builds advertising into the model. It sits by the freeway. It gets people talking. It's interesting. It represents what we do.
Maxims
- The product is not the point. When the goods can't be differentiated, the experience of buying them is where you compete.
- Make the building do the marketing. Infrastructure that is also a landmark spreads its cost across jobs that would otherwise be separate budgets.
- Every advantage gets audited eventually. Family capital, related-party deals, and inherited infrastructure will all be judged by someone who doesn't share your view of them.
- Delight dies in the paperwork. A beautiful checkout doesn't help if the title arrives late.
- Predict the near-death experience. Telling your team in advance that the company will nearly die turns panic into an expected stage.
- Give from your own pocket. Equity transferred from the founder's own holdings carries a message no company grant can.
- Distrust your own excitement. If your instinct is to jump to the next thing, hire people whose job is to stop you.
- Cut the plan, keep the thesis. Overextension and a wrong idea call for different responses, so be sure which one you're dealing with.
- Put money in at the bottom. Buying when you're being written off is the most credible form of conviction available.
- Control concentrates scrutiny. Owning the votes protects you from being overruled, but not from being examined.
