A Lesson in Not Taking Shortcuts
The recent emergence of the modern plug-in electric vehicle has brought more automaker start-ups to the marketplace than we’ve seen in decades, with Fisker Automotive and Tesla Motors the most prominent examples. While there were numerous parallels and connections between the two California-based start-ups early on, the last 12 months have seen a dramatic divergence in their fates. How did it happen?
In hindsight, it seems that Fisker ignored that warning we all heard so many times from our high-school algebra teachers not to skip any steps in the process. If steps are skipped, the chances of messing up are dramatically increased.
Ten years ago Martin Eberhard and Marc Tarpenning launched what would become Tesla Motors. They soon were joined by Elon Musk, who had a big bag of cash from selling his share of PayPal to eBay. The goal was to build a production electric sports car similar on the tzero concept developed by AC Propulsion (ACP). ACP was uninterested in the immense effort required to productionize the wickedly fast and emissions-free roadster, so the trio of tech entrepreneurs took on the challenge.
Since they were starting from scratch in the car business, they opted to adapt a lot of existing technology for the Tesla Roadster. Because there were no lithium-ion cell manufacturers producing cells specifically designed for automotive applications, chief technology officer JB Straubel and his team followed ACP’s lead and licensed some of their technology. They used the same well-proven, if somewhat volatile, lithium-cobalt-oxide cells found in millions of laptop computers.
Despite the batteries’ propensity to occasionally self-immolate, manufacturers such as Panasonic knew how to mass-produce these cells, and the Tesla engineers devised a packaging strategy to keep them cool and monitor their condition.
Instead of designing a platform from scratch, Tesla teamed up with Lotus to adapt the remarkably lightweight yet strong aluminum chassis from the Elise. Constructed from a mix of riveted and bonded aluminum extrusions, stampings, and castings, the Elise architecture was easily stretched to accommodate different needs. In Tesla’s case, the wheelbase was extended to provide a slightly longer cabin and room for the mid-mounted battery pack.
While the production Roadster shared only a handful of parts with the Elise, mutating its DNA allowed Tesla to create a relatively robust debut entry into the marketplace, even if the launch was about two years behind schedule and well over budget.
Starting from scratch (mostly)
Henrik Fisker took a decidedly different approach with the Karma. He launched his company in late 2007 after an unhappy time working with none other than Tesla. Fisker’s design firm had been contracted by Tesla to design its second car, which evolved into the Model S.
No stranger to the auto industry, Fisker had previously served as a designer at BMW and Aston Martin prior to striking out on his own with Fisker Coachbuild. Fisker and his cohorts decided to dive right in and create an all-new vehicle from the wheels up with a new powertrain architecture using many parts from suppliers that were new to the automotive business.
The Karma was designed around a plug-in series hybrid layout similar to what General Motors was developing for the Chevrolet Volt. One of the few major components sourced from a known high-volume supplier was the 2.0-liter turbocharged GM four-cylinder as a range extender. Fisker Automotive investor Quantum Technologies, which had automotive experience mainly as a supplier for low-volume research projects like fuel cell vehicles, supplied the electric motors for the Karma.
Of course the single most important element of any plug-in vehicle is the battery pack. Until recently, no company had any experience making large-scale lithium-ion batteries for high-volume automotive applications. After jumping from Advanced Lithium Power to Enerdel, Fisker finally landed with the also-ran on the Volt program, A123 Systems. A123 had some very promising technology, but its lack of experience in the automotive space was one of the reasons cited by GM when it opted to go with LG Chem for the Volt.
What could possibly go wrong?
So while Tesla used proven battery technology from reliable suppliers in a new application, Fisker went all in with new technology from a new source in a new product from a new company.
A lot could and did go wrong. Basic development work had already begun in early 2006, well before GM showed the Volt concept to the world in January 2007. Within a couple of months, GM had awarded battery development contracts to LG Chem and A123. By the end of October of that year, GM already had the first prototype packs in its lab. By the time the Volt went on sale in late 2010, it had accumulated tens of thousands of hours of charge and discharge cycles on hundreds of batteries in the lab and in vehicles. This is the kind of progress that can only be made by testing many batteries around the clock.
The scale of Tesla’s testing was significantly smaller, but even it had more than three years of experience by the time the Roadster went on sale, and another four years by the time the first Model S was delivered.
Fisker showed the first design mockup of the Karma in January, 2008 in Detroit, and a somewhat more complete version a year later. As late as the 2009 Detroit show, Henrik Fisker was still saying that Karma deliveries would begin by the end of that year, even though no fully functional prototype existed. It wasn’t until mid-2009 that A123 Systems even came up in the conversation. As every automaker (not to mention Boeing with its recent experience with the 787) that has worked with plug-in vehicles will affirm, those batteries require an awful lot of testing in a lot of conditions to verify durability and understand how they work so that they aren’t overcharged or discharged.
There is simply no substitute at this point for the time and effort required to conduct that testing. Even Tesla, which has had a remarkably reliable run so far, has seen issues with some Roadster batteries that got bricked when they were completely discharged and left to sit for a time.
With the apparent indecision about battery suppliers, it was clear to pretty much everyone not affiliated with Fisker that there was no way customers would be getting Karmas for Christmas in 2009. We don’t know what kind of battery was installed when Fisker finally got its first running prototypes in mid-2009, but with the limited resources of a startup, there is no way Fisker was able to do anywhere near the sort of testing with A123 batteries that Tesla almost certainly managed.
Competing business models
Back in early 2007, when Martin Eberhard was still at Tesla’s helm, he explained the strategy he and his partners had devised. Prior to Tesla, most attempts at building electric vehicles had revolved around less expensive commuter cars that would appeal to the hard-core environmental crowd. As any automaker will tell you, it’s really hard to make money on inexpensive cars, and without major cash flow, you can’t sustain a business as capital-intensive as building cars.
From looking at other high-end sports cars, Tesla recognized that customers in that segment were often willing to tolerate some quirks for a great driving experience. Despite a lack of experience in the auto business, the Tesla team was aware it would likely experience some stumbles early on. Since Eberhard and Musk had gotten into this just hoping to build some electric sports cars, this first phase fit together nicely.
Once they got to work on the Roadster, which was always meant as a lower-volume plaything, the plan was to use the revenue to finance development of less-expensive, higher-volume, mainstream models, then gradually moving downmarket. While Tesla certainly struggled and very nearly ran out of cash on several occasions, timely infusions from investors as well as Daimler and later Toyota carried it through.
Despite a lot of doubts from observers and delays to products, Tesla stuck to the plan. While the follow-up to the Roadster originally code-named “White Star” was supposed to be on sale in 2009, Model S deliveries didn’t start until mid-2012. Since then, it has been highly praised and won numerous awards.
The Model S isn’t perfect, and it came to market lacking some of the features found in comparably priced automobiles. Musk took the helm himself in 2008 and has followed the Apple model of selectively editing the feature list and executing the remaining pieces as well as possible. Better to come to market with a shorter checklist done right than a lot of half-finished items.
Just as Tesla released several incremental updates to the Roadster during its four year production life cycle, so is the Model S. Recently, hidden in-car service menus were discovered in the Model S that point to new features such as adaptive cruise control and blind spot detection that will probably arrive before long. This week, Tesla conducted a public demonstration of quick battery swapping, which was promised back in 2009
when the original design concept was shown.
Meanwhile at Fisker, in January 2009 during the height of the great recession, the founder and CEO was claiming the company had 1300 pre-orders for the car and planned to build 15,000 a year by 2010
. As a baseline, in 2012, when the market was already well on its way to recovery, Mercedes-Benz and BMW sold fewer than 12,000 S-Class and 7 Series, and Audi sold half as many A8s, making Fisker’s goals sound even more unrealistic.
There is nothing wrong with your reach exceeding your grasp. However when you have to invest in development and production of something as complex as a car — especially one with so much new technology — it’s better to have more realistic expectations.
Unlike Tesla, which gradually built up the business by selling low-volume Roadsters as well as California zero-emissions-vehicle credits and electric drive systems to Daimler and Toyota, Fisker jumped right into the deep end. It created an all-new vehicle with all-new technology in a segment filled with excellent products for customers who are far less likely to tolerate major bugs.
By the time the first Karmas were put into the hands of paying customers, they proved to be extremely cramped inside, had lower than expected range and fuel efficiency, buggy and slow infotainment systems, and short-circuiting batteries, among other issues. It was a disaster.
Fisker’s approach was a classic case of skipping a couple steps in the algebra solution, dropping a negative sign for a positive and messing up the order of operations. At the end of the day, it just doesn’t add up.
How did it happen?
It probably comes down to money and control.
A few weeks ago, Tesla held a secondary public stock offering and raised more than $1 billion in new cash. Nearly half of that was used to pay off a $465 million Department of Energy loan nine years early. Currently trading at more than $100 per share, Tesla is valued at nearly six times its 2010 IPO price.
Make no mistake, it has been a bumpy ride over the past ten years, and many of the people who have worked for Elon Musk have complaints about him on a personal level. In many respects, Musk is very similar to the late Steve Jobs. He is often described as a workaholic control freak. He is CEO of two companies, Tesla and SpaceX and he is very hands-on at Tesla. Ever since he took over the CEO reins there in 2008, there has been no doubt who is in charge.
Prior to going public, Tesla raised something over $300 million during five funding rounds, with a substantial portion of that coming from Musk’s own pocket. From the very beginning when he joined up with Eberhard and Tarpenning, Musk has had a strong vision of where the company should go and how it should get there. It was a very rational plan and he has stuck to it.
When it comes to engineering, additional resources can help sometimes, but not always. There are times when no matter how much money or how many bodies you throw at a task, the task just takes time to complete properly and can’t be rushed.
Beginning in late 2007, Fisker Automotive raised somewhere around $1.2 billion from its investors, including noted Silicon Valley venture capital firm Kleiner Perkins Caufield & Byers and its managing partner Ray Lane. Lane and Kleiner missed out on the opportunity
to get in early on Tesla.
It will probably some years before we can find out what really happened behind closed doors at Fisker, but we can certainly do some educated speculation now. In business, when there are large sums of money involved, there are almost inevitably equally large egos. There was certainly some bad blood between Elon Musk and Henrik Fisker.
With the amount of money that investors including Ray Lane and actor Leonardo di Caprio had on the line, there was almost certainly tremendous pressure on Fisker to bring the Karma to market as soon as possible to beat Tesla. Given Musk’s accusations of sub-standard work against Fisker, Fisker was no doubt anxious to prove the Tesla CEO wrong on that point as well.
Given the nature of the problems that occurred in the field once Karma deliveries began, it’s almost certain that another year of development and testing and perhaps a slower production ramp-up would have been a huge benefit to customers, and in the long run, to the company. Fisker certainly accomplished a lot going from a design buck to production in three and a half years, but when you’re asking customers to pay $100,000 or more for a car, that’s just not good enough.
In the end, Fisker Automotive proved once again that there are no shortcuts to success in business. If you expect customers to pay big bucks to fund your future, you’d better deliver, and Fisker didn’t. Today Tesla has a market value of $11.8 billion while Bob Lutz and China’s Wanxiang have reportedly bid a mere $20 million to buy the remains of Fisker Automotive.