In January 2002, Elon Musk was basking in the sun on a Rio de Janeiro beach. It was just where you'd expect to find a 30-year-old who was about to see PayPal — the electronic payments company where he was a board member and the largest shareholder — go public. But Musk wasn't flipping idly through People or downing pitchers of caipirinhas. He was engrossed in a different kind of vacation reading: Fundamentals of Rocket Propulsion. Musk, a former physics student, wasn't a rocket scientist. But he was about to become one.
Five months later, Musk used some of his estimated $328 million fortune to fund Space Exploration Technologies (SpaceX), with the ambitious — some might say absurd — goal of building a rocket that would send small payloads into low-Earth orbit at one-tenth the going rate in the United States. Two and a half years and more than $50 million later, the Falcon I, a 60,000-lb. rocket that will deliver a 1,500-lb. payload, is on the verge of its first launch. It if goes up, it will be the first completely privately funded rocket in history to make it into orbit. And even if the first flight fails, SpaceX will have come further, faster, than any other company in the space business. "I think SpaceX has a good shot to revolutionize the industry," says Nathaniel Mass, a senior fellow at Katzenbach Partners LLC, a consulting firm that has studied SpaceX.
Space is all the rage these days, thanks to the success of Burt Rutan and his SpaceShipOne, which in October 2004 claimed the $10 million Ansari X Prize for the first private manned spacecraft to reach suborbital space two times. Well-heeled Internet folks in particular seem drawn to the concept. There's Jeff Bezos, whose mysterious Blue Origin is said to be working on manned spaceflight; John Carmack, cocreator of the video game DOOM, whose Armadillo Aerospace competed for the X Prize; and Paul Allen, who backed Rutan. But the industry is littered with the carcasses of those sure they could break the stranglehold of NASA, Boeing, Lockheed Martin, and others.
To tackle this sphere without years of training or connections is audacious in itself. To proclaim your intent to transform the industry is even more so. But Musk and his team have created an entrepreneurial culture that dares to dream big — a throwback to the anything-is-possible philosophy of the New Economy — while at the same time taking a deeply practical approach to creating a hunk of metal that may hurtle 200 to 500 miles in the air (but may also explode before it ever leaves the ground). Though Musk's goals are ambitious, his approach is radically simple: How can we build a rocket faster, cheaper, and better than ever before? And how can a dream team of great minds avoid the human roadblocks that can doom any working group?
Like Southwest Airlines, Nucor, Dell, and other innovators, SpaceX derives its success from a series of small improvements done on the cheap. There aren't any big R&D labs here. No battalions of PhDs. And no government subsidies. SpaceX is a place where innovation is a state of mind. "SpaceX is not a shiny-thing innovation company," says Josh Olken, senior associate at Katzenbach. "It's about process."
It's a Thursday afternoon at SpaceX's 55,000-square-foot headquarters in El Segundo, the heart of Southern California's aerospace sector, and apart from a few people gathered for a quick snack in the makeshift kitchen, everyone is busy building a rocket. Giant metal shapes and cones are scattered all over the floor. Dominating the room is the first-stage fuel tank, built of aluminum 2219 and laid out horizontally. Inside a plastic tent, a man is welding titanium, one of the skills that SpaceX has decided to bring in-house to improve quality. In the cubicles adjoining the work floor, actual rocket scientists, virtually all men, huddle over computers, discussing the latest results of various tests. The mood is both serious and convivial.
Two rockets are currently in production — Falcon I and Falcon V, the much larger follow-on rocket that will deliver a payload of up to 12,000 lbs. Falcon I will charge customers a fixed rate of $5.9 million to send up a small satellite. That's two-thirds cheaper than the going rate in the United States. Falcon I already has three paying customers, including the government of Malaysia and the U.S. Department of Defense, and Falcon V has one, even before the first Falcon I has launched. There are rumors of another, even larger rocket in the works, referred to in-house as BFR, for Big F — king Rocket.
SpaceX is a creative mix of new school and old school, of dotcom and DEFCON, of 1999 and 1969. On the dotcom side, the kitchen is constantly replenished with cookies, Clif Bars, juice, and chips. Musk's four dogs are regular visitors. There's an air-hockey table and a few Segways strewn about for general use. And there are a lot of geeky-looking guys hard at work, most of whom choose to wear SpaceX T-shirts every day. ("I said, 'Wear your regular clothes, people!' " says Mary Beth Brown, Musk's assistant. "We're not a cult!") SpaceX has much of the feel of a software company about to release a new product. But these guys can't put out a fix if something goes wrong. "You can always patch a software release," says Musk, who, yes, does own two jets and one Russian fighter plane. "It's hard to patch the rocket."
Yet there are lots of things happening at SpaceX that have little to do with the table games or the free snacks. Musk is funding the company himself, and he has challenged his employees to do more with less. By getting people to think creatively about cost, he thinks he can alter the economics of space for good. "Many times we've been asked," he says. " 'If you reduce the cost, don't you reduce reliability?' This is completely ridiculous. A Ferrari is a very expensive car. It is not reliable. But I would bet you 1,000-to-1 that if you bought a Honda Civic that that sucker will not break down in the first year of operation. You can have a cheap car that's reliable, and the same applies to rockets."
To save money, SpaceX's engineers pride themselves on tweaking existing technologies and looking beyond the aerospace world for inspiration. There may not be hundreds of patents in Falcon I, but the scientists don't care. What they want is a rocket that flies. So the main engine isn't a 21st-century design but rather a 1960s-era pintle engine, which has only one fuel injector rather than the "showerhead" or flat-faced injector used in most rockets. It is old, but reliable.
At SpaceX, no money-saving idea is too wacky, so long as it works. Instead of buying a new theodolite, a tool used to align the rocket, an employee bought one on eBay, saving $25,000. When doing research on the fairing, an enormous piece of metal that protects the payload, Chris Thompson, VP of production operations, discovered an old one in an industrial junkyard and had it transported back to the shop for testing.
The company has also borrowed parts and technology from other industries, heresy in such a specialized realm. Falcon I's first stage is partly reusable, and will land in water. Typically, to hire a company specializing in rocket retrieval costs about $250,000. But there are commercial salvage companies accustomed to handling sensitive equipment. A bit of research located Sause Bros., a tug-and-barge company that has agreed to pull in the rocket for $60,000.
In avionics, rather than using an aerospace computer on the rocket, which could cost as much as $1 million, Falcon I will fly with the same kind of computer used in an automatic teller machine for a cost of $5,000. And rather than relying on the costly electronics that NASA and others are wedded to, Falcon I will use an ethernet bus for communication between the rocket's different computers. "I didn't want to invent anything new," says Hans Koenigsmann, SpaceX's VP of avionics and systems.
Once liberated from the constraints of thinking like, well, rocket scientists, SpaceX's employees found inspiration in odd places. While looking for a low-cost way to make tanks, Thompson drove by a milk truck and stopped cold. "Those things are mass produced, there are hundreds on the road, and not a lot of failures," he thought. "Couldn't we use something similar?" He met with the manufacturer and ultimately realized it wouldn't work. But it might have.
Yet SpaceX is not about low cost at any cost; it is the yin and yang of Musk's practicality and his willingness to spend when necessary that gives SpaceX an edge. At SpaceX, VPs don't need to fill out cumbersome requisition forms in triplicate when they need to buy equipment: They are trusted to find the practical, lowest-cost solution, and they just buy it. If it costs more than $5,000, they walk upstairs to Musk's messy cubicle and simply make their case.
"Elon has really opened the checkbook for me," says Bob Reagan, the ponytailed, gravel-voiced VP of manufacturing. "All he ever asks me is, 'What do you need?' " That trust — and the fact that all employees are stockholders — spurs them to try to save Musk money. "It doesn't make any sense to hurt the house," Reagan says. "If we can save a penny, we do."
To build a rocket from scratch in less than three years is no easy task, but it helps if speed is a core value. Yet it is not speed for speed's sake: Rather than toiling for years on a part until it is perfect, SpaceX's approach is to build as quickly as possible and then, in the words of Tom Mueller, VP of propulsion development, "test the crap out of it." This commitment to fast prototyping and testing has been a constant theme through Falcon I's development. For example, the use of carbon composite instead of aluminum in some parts allows them to be assembled in three days instead of two to three weeks.
Getting something out there fast inevitably leads to failures — but that's to be expected at SpaceX. When something goes wrong, "first of all, there's a good lashing," jokes Musk. Getting serious, he says, "There's a silly notion that failure's not an option at NASA. Failure is an option here. If things are not failing, you are not innovating enough." In fact, says Kevin Brogan, a senior design engineer, "the first time we had a major engine failure Elon was kind of excited. It gave him some street cred." Told this story, Musk smiles wanly. "If I had the option of not having it blow up, I'd rather not," he says. "But it was pretty cool."
Sometimes, the SpaceX team has found, the quickest way to get a part or material to the standards required is to make it themselves. Musk hired Reagan to build a $10 million in-house machine shop where parts are made quicker and cheaper — like a $25,000 valve that now costs SpaceX $11,000. The investment may or may not ultimately pay off, but for Musk, it's where innovation comes out of necessity. "The pace of execution is really the fundamental [advantage]," he says. "I consider it almost synonymous with competitiveness."
The third part of SpaceX's equation is the "better" part, and that is where the team comes in. Although SpaceX is growing at 15% a month — it's now at 88 employees — Musk still sees its limited size as a plus. "A small group of very technically strong people will always beat a large group of moderately strong people," he says. It's a lesson he learned at PayPal, where 30 engineers went up against Citibank, which was trying to develop a competing product.
Back in 2002, while Musk was teaching himself propulsion, he was also bringing together the smartest minds in aerospace. In part from the contacts he made through these "feasibility studies," Musk cherry-picked the best, brightest, and most bored from other companies. His selling point was that they'd have the freedom to actually do their job — build a rocket — rather than sitting in daylong meetings, waiting months for a parts request to wend its way through a bureaucracy, or fending off internal political attacks. "This is the first job where I'm not bored," says Brogan, who worked at TRW for five years before coming to SpaceX. At TRW, Brogan says, he had to toil for years on a program that everyone knew was doomed because of a change in government priorities, but had already been paid for. "That program really destroyed my soul," he says.
At SpaceX, there are only five employees who don't have direct responsibility for a particular component or technology, and that personal connection is a key motivator. "I have friends who are happy at their jobs and who ask me why I work so much," says Brian Bjelde, a young avionics engineer formerly at NASA. "They're working on some component that may or may not fly. They'll analyze it for four years, and it may go on a spacecraft that launches in 2015. That's just not me." To encourage openness, Musk hosts a regular Friday lunch where anyone can ask questions about where SpaceX is going. "Here, it's fourth down and you've got three yards to go," says Brogan. "At TRW, we didn't know if there was even a football."
SpaceX operates as a nearly horizontal organization. Musk is unquestionably the boss, and there are seven vice presidents, all of whom are hands-on specialists in things such as propulsion or avionics. There are no paper-pushing managers and no organization charts. Until recently, when the company brought in a head of human resources, Musk and the VPs interviewed every single job candidate, and everyone had veto power. While pure intelligence is an absolute must — the team delights in coming up with mental challenges during interviews — it's equally important that candidates disdain the political and embrace SpaceX's underdog culture. "I think it's really unacceptable here for anyone to bear a grudge," Musk says. "It's a showstopper issue." So far, the approach to hiring has worked well. In over two years, one person returned to school, two have been fired, and one person was recalled to a previous job.
If you'd never met Elon Musk, it would be easy to dismiss him as just another dotcom dilettante, a young buck with great timing who grew up in the thrall of Star Wars and now has the scrilla to build his own space playground. But you'd be dead wrong. Musk, a tall blond South African native with an elfin look, a deliberate way of speaking, and a very dry sense of humor, has been trying to shake things up since childhood. At 12, he sold the code for a video game he designed; at 17, he moved to Canada for college; he later transferred to the University of Pennsylvania for a physics and business degree, and dropped out of graduate studies at Stanford on his first day there in 1995, convinced that the Internet was going to be huge. Unable to find a job, he started Zip2 Corp., which helped media companies put content on their Web sites, at the age of 23. In 1999, Musk sold it to Compaq for $300 million, netting more than $20 million.
Next came X.com, a wildly ambitious attempt to reorder the world's financial system. "He's always been a sort of endearing mixture of flightiness and practicality," says Michael Moritz, a venture capitalist at Sequoia Capital Partners and an X.com investor. It turned out that X.com's most successful concept was a secure online payment system. It bought Confinity, a competitor with a similar technology called PayPal, in 2000, and took the name. Musk ran PayPal for several months before leaving, but remained the largest shareholder. In July 2002, eBay bought it for $1.5 billion, lining Musk's already deep pockets with another $165 million in stock.
Next came the final frontier. Distressed to find out that NASA had no defined plan for sending humans to Mars, he began researching the industry and realized that aerospace defied Moore's Law: Instead of access to space getting cheaper over time, it had gotten costlier, without getting more reliable. "We sleep easy knowing that next year's software will be better than this year's," he says. "Rockets' [cost] actually gets progressively worse every year."
Musk smelled opportunity. Why, he wondered, were rockets so expensive? He decided to use some of his cash to bring together the world's top space experts to find out in early 2002. The group met at the Hyatt near the Los Angeles airport over several Saturdays, discussing whether it was possible to build a better rocket. "He asked, 'How much cheaper?' " says Mueller. "A factor of 3? I said a factor of 10, if you could keep the government off your back."
At the meetings, the group helped Musk begin designing his better, faster, cheaper rocket. Many of the initial decisions were made by Musk, who has surprised people with his technical know-how (and who is also the company's chief technology officer). "I am very impressed with how fast he came up to speed," says Philip McAllister, director at Futron Corp., an aerospace consulting firm. "He seems to have gotten a good grasp on not only the technical hardware but how the business works."
Although everyone at SpaceX is working on the Falcon I, Musk isn't shy about proclaiming a much bigger goal. "We work harder than other companies," a holiday party flyer reminds staffers, "because we are quite literally shifting the course of history to open space for humanity." Ultimately, says Musk, he wants to make space travel attainable by ordinary people. "I think human exploration of space is very important," Musk says. "Certainly, from a survival standpoint, the probability of living longer is much greater if we're on more than one planet."
But if Musk seems to veer off into Twilight Zone territory here, he just as quickly circles back to hard reality. After all, he has spent more than $50 million of his own money so far and says he's willing to go as high as $100 million — maybe higher. "I tell my wife that if this fails, we'll have to move into my parents' basement," he laughs.
He acknowledges that the failure rate for a first-time launch is generally high, and says that he's willing to fund as many as three of them before giving up. Still, he professes a giddy optimism that the maiden launch will be successful. "Touch the bloody wood," Musk says, putting both hands down on the table. "I think we've got well over a 90% chance of success in the first launch."
Can Musk and his merry band pull it off? Not everyone is rooting for them. Earlier this year, Northrop Grumman sued SpaceX over the pintle engine, which was developed by TRW, now owned by Northrop. Northrop claims that SpaceX violated trade secrets, but SpaceX has countersued, claiming that Northrop is trying to extend a patent that has expired with its trade-secrets argument. The case is in discovery, and could put a freeze on the Falcon I, although no one seems concerned. There's also the prospect that the government, still the biggest source of business in this market, will continue to favor the big defense companies. "They absolutely can be successful, but the biggest risk is government," says Andrew Beal, CEO of Beal Bank, who spent more than $100 million on an earlier effort to build a private rocket. "My advice is to be careful."
Others wonder whether there is actually a market for such small payloads, since until now the cost of the launch has made it pointless to send them up. Musk believes, in a classic Field of Dreams strategy, that if you make the launch cost-effective, the business will appear. "He can make something happen," says John Garvey, CEO of Garvey Spacecraft Corp., a small aerospace tech-development company. "But I have questions as to whether the market can make something happen."
In November, problems with the injector caused another short delay in the launch, moving it to late February, over a year behind the original schedule. The news was certainly disappointing. But there was no overt gnashing of teeth or pulling of hair over the development, which Musk disclosed over pizza at the regular Friday meeting. Everyone absorbed it, rolled up their sleeves, and got back to work. They were building a rocket, after all, not to mention trying to change the world.
Jennifer Reingold is a Fast Company senior writer.
A version of this article appeared in the February 2005 issue of Fast Company magazine.