by Ashlee Vance
Tesla also faced issues abroad. The company had decided to send a team of its youngest, most energetic engineers to Thailand to set up a battery factory. Tesla partnered with an enthusiastic although not totally capable manufacturing partner. The Tesla engineers had been told that they could fly over and manage the construction of a state-of-the-art battery factory. Instead of a factory, they found a concrete slab with posts holding up a roof. The building was about a three-hour drive south from Bangkok, and had been left mostly open like many of the other factories because of the incredible heat. The other manufacturing operations dealt with making stoves, tires, and commodities that could withstand the elements. Tesla had sensitive batteries and electronics, and like parts of the Falcon 1, they’d be chewed up by the salty, humid conditions. Eventually, Tesla’s partner paid about $75,000 to put in drywall, coat the floor, and create storage rooms with temperature controls. Tesla’s engineers ended up working maddening hours trying to train the Thai workers on how to handle the electronics properly. The development of the battery technology, which had once moved along at a rapid pace, slowed to a crawl.
The battery factory was one part of a supply chain that stretched across the globe, adding cost and delays to the Roadster production. Body panels for the car were to be made in France, while the motors were to come from Taiwan. Tesla planned on buying battery cells in China and shipping them to Thailand to turn the piece parts into battery packs. The battery packs, which had to be stored for a minimal amount of time to avoid degradation, would then be taken to port and shipped to England, where they needed to clear customs. Tesla then planned for Lotus to build the body of the car, attach the battery packs, and ship the Roadsters by boat around Cape Horn to Los Angeles. In that scenario, Tesla would have paid for the bulk of the car and had no chance to recognize revenue on the parts until six to nine months had passed. “The idea was to get to Asia, get things done fast and cheap, and make money on the car,” said Forrest North, one of the engineers sent to Thailand. “What we found out was that for really complicated things, you can do the work cheaper here and have less delays and less problems.” When some new hires came on, they were horrified to discover just how haphazard Tesla’s plan appeared. Ryan Popple, who had spent four years in the army and then gotten an MBA from Harvard, arrived at Tesla as a director of finance meant to prep the company to go public. After examining the company’s books early in his tenure, Popple asked the manufacturing and operations head exactly how he would get the car made. “He said, ‘Well, we will decide we’re going into production and then a miracle is going to happen,’” Popple said.
As word of the manufacturing issues reached Musk, he became very concerned about the way Eberhard had run the company and called in a fixer to address the situation. One of Tesla’s investors was Valor Equity, a Chicago-based investment firm that specialized in fine-tuning manufacturing operations. The company had been drawn to Tesla’s battery and powertrain technology and calculated that even if Tesla failed to sell many cars, the big automakers would end up wanting to buy its intellectual property. To protect its investment, Valor sent in Tim Watkins, its managing director of operations, and he soon reached some horrific conclusions.
Watkins is a Brit with degrees in industrial robotics and electrical engineering. He’s built up a reputation as an ingenious solver of problems. While doing work in Switzerland, for example, Watkins found a way to get around the country’s rigid labor laws that limit the hours employees can work, by automating a metal stamping factory so that it could run twenty-four hours per day instead of sixteen hours like the factories or rivals. Watkins is also known for keeping his ponytail in place with a black scrunchie, wearing a black leather jacket, and toting a black fanny pack everywhere he goes. The fanny pack has his passport, checkbook, earplugs, sunscreen, food, and an assortment of other necessities. “It’s full of the everyday things I need to survive,” said Watkins. “If I walk ten feet away from this thing, I sense it.” While a bit eccentric, Watkins was thorough and spent weeks talking to employees and analyzing every part of Tesla’s supply chain to figure out how much it cost to make the Roadster.
Tesla had done a decent job of keeping its employee costs down. It hired the kid fresh out of Stanford for $45,000 rather than the proven guy who probably didn’t want to work that hard anyway for $120,000. But when it came to equipment and materials, Tesla was a spending horror show. No one liked using the company’s software that tracked the bill of materials. So some people used it, and some people didn’t. Those that did use it often made huge errors. They would take the cost of a part from the prototype cars and then estimate how much of a discount they expected when buying that part in bulk, rather than actually negotiating to find a viable price. At one point, the software declared that each Roadster should cost about $68,000, which would leave Tesla making about $30,000 per vehicle. Everyone knew the figure was wrong, but it got reported to the board anyway.
Around the middle of 2007, Watkins came to Musk with his findings. Musk was prepared for a high figure but felt confident that the price of the car would come down significantly over time as Tesla ironed out its manufacturing process and increased its sales. “That’s when Tim told me it was really bad news,” Musk said. It looked like each Roadster could cost up to $200,000 to make, and Tesla planned to sell the car for only around $85,000. “Even in full production, they would have been like $170,000 or something insane,” Musk said. “Of course, it didn’t much matter because about a third of the cars didn’t flat-out fucking work.”
Eberhard made attempts to pull his team out of this mess. He’d gone to see a speech in which the famous venture capitalist John Doerr, who became a major investor in green technology companies, declared that he would devote his time and money to trying to save the Earth from global warming because he owed such an effort to his children. Eberhard promptly returned to the Tesla building and ginned up a similar speech. In front of about a hundred people, Eberhard had a picture of his young daughter projected onto the wall of the main workshop. He asked the Tesla engineers why he had put that picture up. One of them guessed that it was because people like his daughter would drive the car. To which Eberhard replied, “No. We are building this because by the time she is old enough to drive she will know a car as something completely different to how we know it today, just like you don’t think of a phone as a thing on the wall with a cord on it. It’s this future that depends on you.” Eberhard then thanked some of the key engineers and called out their efforts in public. Many of the engineers had been pulling all-nighters on a regular basis and Eberhard’s show boosted morale. “We were all working ourselves to the point of exhaustion,” said David Vespremi, a former Tesla spokesman. “Then came this profound moment where we were reminded that building the car was not about getting to an IPO or selling it to a bunch of rich dudes but because it might change what a car is.”
These victories, though, were not enough to overcome the feeling shared by many of the Tesla engineers that Eberhard had reached the end of his abilities as a CEO. The company veterans had always admired Eberhard’s engineering smarts and continued to do so. Eberhard, in fact, had turned Tesla into a cult of engineering. Regrettably, other parts of the company had been neglected, and people doubted Eberhard’s ability to take the company from the R&D stage to production. The ridiculous cost of the car, the transmission, the ineffective suppliers were crippling Tesla. And, as the company started to miss its delivery dates, many of the once-fanatical consumers who had made their large up-front payments turned on Tesla and Eberhard. “We saw the writing on the wall,” Lyons said. “Everyone knew that the person who starts a company is not necessarily the right person to lead it in the long term, but whenever that is the case, it’s not easy.”
Eberhard and Musk had battled for years over some of the design points on the car. But for the most part, they had gotten along well enough. Neither man suffered fools. And they certainly shared many of the same visions for the battery technology
and what it could mean to the world. What their relationship could not survive were the cost figures for the Roadster unearthed by Watkins. It looked to Musk as if Eberhard had grossly mismanaged the company by allowing the parts costs to soar so high. Then, as Musk saw it, Eberhard failed to disclose the severity of the situation to the board. While on his way to give a talk to the Motor Press Guild in Los Angeles, Eberhard received a call from Musk and in a brief, uncomfortable chat learned that he would be replaced as CEO.
In August 2007, Tesla’s board demoted Eberhard and named him president of technology, which only exacerbated the company’s issues. “Martin was so bitter and disruptive,” Straubel said. “I remember him running around the office and sowing discontent, as we’re trying to finish the car and are running out of money and everything is at knife’s edge.” As Eberhard saw it, other people at Tesla had foisted a wonky finance software application on him that made it tricky to accurately track costs. He contended that the delays and cost increases were partly due to the requests of other members of the management team and that he’d been up front with the board about the issues. Beyond that, he thought Watkins had made the situation out to be worse than it really was. Start-ups in Silicon Valley view mayhem as standard operating procedure. “Valor was used to dealing with older companies,” Eberhard said. “They found chaos and weren’t used to it. This was the chaos of a start-up.” Eberhard had also already been asking Tesla’s board to replace him as CEO and find someone with more manufacturing experience.
A few months passed, and Eberhard remained pissed-off. Many of the Tesla employees felt like they were caught in the middle of a divorce and had to pick their parent—Eberhard or Musk. By the time December arrived, the situation was untenable, and Eberhard left the company altogether. Tesla said in a statement that Eberhard had been offered a position on its advisory board, although he denied that. “I am no longer with Tesla Motors—neither on its board of directors nor an employee of any sort,” Eberhard said in a statement at the time. “I’m not happy with the way I was treated.” Musk sent a note to a Silicon Valley newspaper saying, “I’m sorry that it came to this and wish it were not so. It was not a question of personality differences, as the decision to have Martin transition to an advisory role was unanimous among the board. Tesla has operational problems that need to be solved and if the board thought there was any way that Martin could be part of the solution, then he would still be an employee of the company.”9 These statements were the start of a war that would drag on between the two men in public for years and that in many ways continues to the present day.
As 2007 played out, the problems mounted for Tesla. The carbon-fiber body that looked so good turned out to be a huge pain to paint, and Tesla had to cycle through a couple of companies to find one that could do the work well. Sometimes there were faults in the battery pack. The motor short-circuited now and again. The body panels had visible gaps. The company also had to face up to the reality that a two-speed transmission was not going to happen. In order for the Roadster to achieve its flashy zero-to-60 times with a single-speed transmission, Tesla’s engineers had to redesign the car’s motor and inverter and shave off some weight. “We essentially had to do a complete reboot,” Musk said. “That was terrible.”
After Eberhard was removed as CEO, Tesla’s board tapped Michael Marks as its interim chief. Marks had run Flextronics, an enormous electronics supplier, and had deep experience with complex manufacturing operations and logistics issues. Marks began interrogating various groups at the company to try to figure out their problems and to prioritize the issues plaguing the Roadster. He also put in some basic rules like making sure that people all showed up at work at the same time to establish a baseline of productivity—a tricky ask in Silicon Valley’s work anywhere, anytime culture. All of these moves were part of the Marks List, a 10-point, 100-day plan that included eliminating all faults in the battery packs, getting gaps between body parts to less than 40 mm, and booking a specified number of reservations. “Martin had been falling apart and lacked a lot of the discipline key for a manager,” Straubel said. “Michael came in and evaluated the mess and was a bullshit filter. He didn’t really have a dog in the fight and could say, ‘I don’t care what you think or what you think. This is what we should do.’” For a while, Marks’s strategy worked, and the engineers at Tesla could once again focus on building the Roadster rather than on internal politics. But then Marks’s vision for the company began to diverge from Musk’s.
By this time, Tesla had moved into a larger facility at 1050 Bing Street in San Carlos. The bigger building allowed Tesla to bring the battery work back in-house from Asia and for it to do some of the Roadster manufacturing, alleviating the supply chain issues. Tesla was maturing as a car company, although its wild-child start-up streak remained well intact. While strolling around the factory one day, Marks saw a Smart car from Daimler on a lift. Musk and Straubel had a small side project going on around the Smart car to see what it might be like as an electric vehicle. “Michael didn’t know about it, and he’s like, ‘Who is the CEO here?’” said Lyons. (The work on the Smart car eventually led to Daimler buying a 10 percent stake in Tesla.)
Marks’s inclination was to try to package Tesla as an asset that could be sold to a larger car company. It was a perfectly reasonable plan. While running Flextronics, Marks had overseen a vast, global supply chain and knew the difficulties of manufacturing intimately. Tesla must have looked borderline hopeless to him at this point. The company could not make its one product well, was poised to hemorrhage money, and had missed a string of delivery deadlines and yet its engineers were still off doing side experiments. Making Tesla look as pretty as possible for a suitor was the rational thing to do.
In just about every other case, Marks would be thanked for his decisive plan of action and saving the company’s investors from a big loss. But Musk had little interest in polishing up Tesla’s assets for the highest bidder. He’d started the company to put a dent in the automotive industry and force people to rethink electric cars. Instead of doing the fashionable Silicon Valley thing of “pivoting” toward a new idea or plan, Musk would dig in deeper. “The product was late and over budget and everything was wrong, but Elon didn’t want anything to do with those plans to either sell the whole company or lose control through a partnership,” Straubel said. “So, Elon decided to double down.”
On December 3, 2007, Ze’ev Drori replaced Marks as CEO. Drori had experience in Silicon Valley starting a company that made computer memory and selling it to the chipmaker Advanced Micro Devices. Drori was not Musk’s first pick—a top choice had turned down the job because he didn’t want to move from the East Coast—and did not inspire much enthusiasm from the Tesla employees. Drori had about fifteen years on the youngest Tesla worker and no connection to this group bonded by suffering and toil. He came to be seen more as an executor of Musk’s wishes than as a commanding, independent CEO.
Musk began making more public gestures to mitigate the bad press around Tesla. He issued statements and did interviews, promising that the Roadster would ship to customers in early 2008. He began talking up a car code-named WhiteStar—the Roadster had been code-named DarkStar—that would be a sedan possibly priced around $50,000, and a new factory to build the machine. “Given the recent management changes, some reassurances are in order regarding Tesla Motors’ future plans,” Musk wrote in a blog post. “The near term message is simple and unequivocal—we are going to deliver a great sports car next year that customers will love driving. . . . My car, production VIN 1, is already off the production line in the UK and final preparations are being made for importation.” Tesla held a series of town hall meetings with customers where it tried to fess up to its problems in the open, and it started building some showrooms for its car. Vince Sollitto, the former PayPal executive, visited the Menlo Park showroom and found Musk complaining about the public relations issues but clearly inspired by the product Tesla was building. “His demeanor changed
the moment we got to this display of the motor,” Sollitto said. Dressed in a leather jacket and slacks, Musk started talking about the motor’s properties and then put on a performance worthy of a carnival strongman by lifting the hundred-or-so-pound hunk of metal. “He picks this thing up and wedges it between his two palms,” Sollitto said. “He’s holding it, and he’s shaking and beads of sweat are forming on his forehead. It wasn’t so much a display of strength as a physical demonstration of the beauty of the product.” While the customers complained a lot about the delays, they seemed to sense this passion from Musk and share his enthusiasm for the product. Only a handful of customers asked for their prepayments back.
Tesla employees soon got to witness the same Musk that SpaceX employees had seen for years. When an issue like the Roadster’s faulty carbon-fiber body panels cropped up, Musk dealt with it directly. He flew to England in his jet to pick up some new manufacturing tools for the body panels and personally delivered them to a factory in France to ensure that the Roadster stayed on its production schedule. The days of people being ambiguous about the Roadster’s manufacturing costs were gone as well. “Elon got fired up and said we were going to do this intense cost-down program,” said Popple. “He gave a speech, saying we would work on Saturdays and Sundays and sleep under desks until it got done. Someone pushed back from the table and argued that everyone had been working so hard just to get the car done, and they were ready for a break and to see their families. Elon said, ‘I would tell those people they will get to see their families a lot when we go bankrupt.’ I was like, ‘Wow,’ but I got it. I had come out of a military culture, and you just have to make your objective happen.” Employees were required to meet at 7 A.M. every Thursday morning for bill-of-materials updates. They had to know the price of every part and have a cogent plan for getting parts cheaper. If the motor cost $6,500 a pop at the end of December, Musk wanted it to cost $3,800 by April. The costs were plotted and analyzed each month. “If you started falling behind, there was hell to pay,” Popple said. “Everyone could see it, and people lost their jobs when they didn’t deliver. Elon has a mind that’s a bit like a calculator. If you put a number on the projector that does not make sense, he will spot it. He doesn’t miss details.” Popple found Musk’s style aggressive, but he liked that Musk would listen to a well-argued, analytical point and often change his mind if given a good enough reason. “Some people thought Elon was too tough or hot-tempered or tyrannical,” Popple said. “But these were hard times, and those of us close to the operational realities of the company knew it. I appreciated that he didn’t sugarcoat things.”