Recently, Business Insider published a fairly thoughtful, historically informed, but ultimately inaccurate opinion piece saying that GM just hadn’t bothered crushing Tesla yet, but was about to. In other words, it looked backward well, but forward poorly.
The part of the article by Matthew DeBord where he veered sharply away from useful commentary was the following statement:
“In the past two years, GM has put the pedal to the metal on everything Tesla pioneered and is now well ahead of the Palo Alto upstart on everything except Tesla’s core business, luxury EVs that sell for $100,000.”
This is an interesting point of view. It shows a deep misunderstanding of Tesla’s competitive differentiation, which isn’t unusual for people who spend a lot of time covering the legacy manufacturers. I’m sure, for example, that his book Return to Glory is a great read about Ford, a legacy brand that’s in deep trouble with the disruptions facing the automotive industry. But DeBord doesn’t cover innovation, he covers history, culture, and driving around wine country. He’s well positioned to articulate Ford and GM’s legacy, but not Tesla’s innovation.
I’ll reiterate the primary points DeBord missed to assist in understanding why GM is going to take a long time to catch up if it ever does, and why it will have to seriously cannibalize its own business to do so. Fundamentally, GM has to compete with itself more than Tesla, and this is a major issue.
Let’s start with the competitive advantages of the Tesla Model 3 over the similarly priced Chevy Bolt. This radar graph I developed is based on my assessment of competitive differentiators. Not all of these are created equal, but many of them are significant game changers. Let’s pick just three to explore for this purpose.
Sales channel as a major problem is poorly understood by legacy automotive journalists and many executives in the automotive industry. They tend to think of this business model as an advantage. However, dealerships benefit most from rapid sales, as they make the majority of their profit on long-term servicing of the vehicle. This is highly problematic for influencing independent dealers to sell electric cars since they have to spend a lot more time educating customers and helping them over range anxiety during the sales process. It’s a lot faster to sell an internal combustion car right now than an electric one unless a highly motivated buyer walks in and demands it. This is clearly seen in the data from Nissan’s Leaf sales and GM’s Bolt sales.
The other side of this equation is that each sale of an electric car instead of an internal combustion car reduces the long-term revenue of the dealership through maintenance. The combination is why smart people are selling dealerships to big aggregators who can find a way to squeeze some profits out of poorer revenues by regional servicing, closing less profitable dealerships and other efficiency measures.
Charging speed and charging points combine for another major difference. The Bolt doesn’t do high-speed charging and there is no intercity charging network for Bolt drivers. They are fundamentally limited to being city cars except for EV enthusiasts like the ones who did long trips in Nissan Leafs. GM is betting on someone else building charging stations, and isn’t building high-speed charging into its cars yet. This isn’t exactly exceeding Tesla’s differentiation or doing anything better than Tesla. Maybe he needs to look at the number of Supercharger stations between major cities worldwide to gain perspective. [Editor’s note: For all practical purposes today, the Bolt has a max charging capacity of 50 kW if the buyer chooses that option. Even if it had a higher charging capacity, there are almost no non-Tesla charging stations that reach 100 kW, with almost all US fast chargers maxing out at 25–50 kW. Tesla’s Superchargers provide up to 120 kW consistently and broadly. Also see: Tesla Supercharger Network 2018.]
I’ll lump together over-the-air-updates and secondary controls being virtualized. Many Chevy Bolts still can’t get over-the-air-updates and Chevy doesn’t have a culture of updates like that. Tesla’s updates are things of legend, with sometimes 10%+ performance improvements. No one has to stick a USB key in or drive to a service station for the virtual problems. And that’s why secondary controls is lumped in with this. The Tesla Model 3 has the absolute bare minimum of knobs, dials, buttons, and switches that are operated manually. Almost everything is virtualized. That means the user experience will continue to improve, while the knobs, dials, buttons, and switches on the Bolt will do what mechanical things do, which is to degrade. [Editor’s note: For more depth on the over-the-air update and overall software advantages, see the presentation from George Hotz at Autonomy 2017 and his Q&A with CleanTechnica there: Geohot: Tesla Autopilot = Apple iOS, Comma.ai = Android (CleanTechnica Exclusive).]
The Chevy Bolt is outselling the Tesla Model 3 due to lack of Tesla Model 3s, not because the Chevy Bolt is particularly great. The 450,000+ lineup for the Model 3 is clear proof of that. It’s selling a reasonable number compared to Model S because it’s a lot cheaper and there’s significant pent up demand that legacy manufacturers have been unwilling and unable to meet in that price point.
Let’s talk about autonomous driving. Here’s another case where DeBord accepts press releases about testing applications from one organization while ignoring the real-world advantages of the other.
He has no background in robotics, machine learning, sensors, or any of the other disciplines necessary to form a technical opinion on the subject. He couldn’t be expected to know about subsumption or Lidar vs radar with digital processing. He’s not that guy.
But he could be expected to look at miles under wheels and how many cars actually have sensors built in vs as extras. Tesla is well into the billions of miles with its autonomous tech under control in real-world conditions in multiple parts of the world. GM is orders of magnitude behind that. Its request to offer a Bolt without a steering wheel in 2019 is deeply unlikely to be accepted, and there’s little reason to believe that in real-world conditions, its technology is actually superior. And, of course, every Tesla has the hardware and computing power built in to achieve full autonomy. Bolts and other GM cars? Most of them in the USA will have automatic emergency braking in 2022, along with almost every other car sold per an agreement the NHTSA hammered out.
I won’t bother getting into the multiple back-to-back tests which found Tesla’s implementation of autonomous features to be much better than those of any other manufacturers. I won’t bother to get into the rapid iterations of Tesla versus the sluggishness of GM’s movement.
Tesla is delivering real-world advanced autonomous features and regularly upgrading them. Those features are in continuous use by drivers. GM isn’t far ahead, it’s struggling to differentiate itself with a marketing release.
Let’s just run the numbers, something DeBord didn’t bother to do. He’s not a math guy or a models guy, so that’s understandable. But it’s not rocket science to look at the relative numbers of pure electric cars Tesla has sold vs GM, look at the competitive landscape, and realize that GM will take a long time to catch up to Tesla under any realistic perspective.
Instead, DeBord asserts that Tesla can’t make cars at scale, as if the relatively temporary challenges with Model 3 production are a foregone conclusion to continue, or as if GM will magically overcome its internal barriers to cannibalizing its own product lines, or as if GM, which is still saying in 2018 that it will deliver hydrogen cars, will be able to deliver a lot of pure electric cars. It’s not like other car companies don’t have teething problems with vehicle launches, but none of them have the scrutiny Tesla does.
The automotive industry is in the throes of a period of major disruptive innovation. It’s pretty clear that DeBord hasn’t read Christensen and Raynor on innovation and, once again, that’s okay. It’s not his patch. I’m sure I’d enjoy reading his books and for anyone who wants to look back with nostalgia written by a decent writer, I’m pretty sure you couldn’t go wrong. But that doesn’t mean you want him giving you guideposts to the future.
DeBord is looking in the rear-view mirror, not out the windshield at the highway that’s coming. I really hope he has a Tesla doing that for him.