X
Innovation

Build less, license more: Tesla needs to be an ARM, not a half-baked Apple

If Tesla wants to scale, then it has to decide what it really wants to be when it grows up.
Written by Jason Perlow, Senior Contributing Writer

Tesla, the electric vehicle automotive manufacturer, has been having a rough couple of weeks. The company recently announced plans to shutter all of its brick-and-mortar stores, and to do only online sales. And there have been a number of high-profile crashes reported in the news, including two very recently in Florida, which prompted federal agencies NTSB and NHTSA to investigate the cars' driver assistance technology, in particular the autopilot systems.

Also: Tesla's new Model Y will be revealed on March 14 CNET

Additionally, the Tesla Model 3, the company's volume production sedan -- which recently became available in a $35,000 entry-level model after years of waiting -- had its Consumer Reports recommendation rescinded after numerous customer reports of reliability issues related to hardware, paint, and trim. 

This is not to say that all Tesla cars are bad, that the technology is all bad, or that the company is incapable of producing a reliable automobile. In fact, I know a number of owners of upper-end Tesla Model S and Model X vehicles who are extremely happy with their purchases.

They should be, because they've spent well over $75,000 and $100,000 for some of their cars.

Also: How to steal a Tesla Model S in seconds

My question is: Can the company actually scale its production to levels of the largest automakers in the world? It's one thing to make a boutique, high-performance luxury car in the 5,000-10,000 unit range per year, as the company did for the first several years of its existence. 

The People's EV

It's a whole other thing to produce a high-volume, electric equivalent of the 1938 Volkswagen -- the people's EV -- which the Model 3 is trying to be.

Tesla is trying to be Bugatti, Audi, Porsche, and Volkswagen. That is fine if you are Volkswagen, with massive amounts of factory infrastructure and a massive supply chain to match, and can produce several brands of cars  with a wide range of different models and different price points -- but Tesla is no Volkswagen. 

Frankly, even Volkswagen doesn't want to be Volkswagen.

Tesla is a very small automotive and battery technology company -- with some very interesting and valuable intellectual property -- that is running on the ego of Elon Musk, who seems to be taking all his direction from the classic Steve Jobs playbook.

Tesla is not Apple

The problem: Not only is Tesla no Volkswagen, but it is also no Apple. To Cupertino's credit, this is a company that has developed not just unique intellectual property and amazing industrial design, but also a entire ecosystem of applications, services, retail store infrastructure, a huge network of accessory and retail partners, and is probably next to Coca-Cola the most recognizable product brand in the world.

It has taken Apple the greater part of a decade to re-invent itself from being the Mac company to the iOS/iPhone company. And, yes, it likely has reached the apex of its growth in that sector. But what it has achieved is extremely impressive, because it has managed to create something without needing to license anything to anyone else in order to make money. 

Like the Mac, you can't run iOS on another OEM's hardware, like the ones that run Android. Their A-series SoCs only run on their hardware, nobody else's. You can't legally run MacOS X on an Intel-based PC, although a small group of dedicated "hackintoshers" do. 

The only way you can run applications on Apple's iOS hardware is through the company's App Store -- third-party, out-of-band installs, such as those permissible on Android, are not allowed. The only way you can do that is through "jailbreaking" or "rooting," which falls under the auspices of warranty invalidation, and then as an end-user you are on your own.

It's a tightly controlled "walled garden." And while I and others have complained about this for many years, it is the primary ideological differentiation that distinguishes Apple from the Android, Linux, and Windows ecosystems. 

Also: What is the Tesla Semi? Everything you need to know

But nobody can say with a straight face that this does not work for Apple. Given market fluctuations in its stock, it is a $800 billion to $1 trillion firm, with a gigantic base of loyal customers.

A sealed, closed box 

Tesla -- despite trying to be Apple -- is not Apple. Unlike Apple, it doesn't have channel sales partners. And even though, as a consumer, you can access the repair manuals for a fee, it is extremely difficult to self-repair the cars when they go out of warranty; only certified body shops that can easily get the parts and other knockdown kits have access to what is needed in terms of diagnostic tools and support. (Although it hasn't stopped a bunch of people from becoming their own rogue support network.) 

Third-party Tesla parts are hard to come by and access to the repair information is expensive. Twenty four-hour access is $100, monthly is $350, and a yearly subscription is $3,000. Certified body shops get it for free, but there are considerable technical and competency requirements for becoming certified. Your average mom and pop garage in rural America can't put together the capital for becoming one of Tesla's elite maintenance shops -- they will be able to work on these cars best effort, if that.

Tesla is the most sealed, closed box you can possibly think of when it comes to cars. It out-Apples Cupertino in its own arrogance, which is a pretty damn amazing feat.

Time to grow up

If Tesla wants to scale, then it has to decide what it really wants to be when it grows up. Maybe it can be an Apple someday, when it reaches a $200 billion valuation. But even at its current $50 billion market cap -- which is vastly over-inflated -- it is a long, long way from that.

Also: Tesla sued: Woman wants $300k for crashing on Autopilot

But until it can actually perform like the company it wants to be, then it needs to find ways of getting its technology into as many hands as possible. That means licensing its technology to large automakers such as the Volkswagens, the GMs, the Hyundais, the Toyotas, and the Nissans of the world. 

Tesla has drivetrains, motors, software, and battery tech, all of which it can license to and partner with other companies that have the resources to provide support infrastructure, sales channels, and brand diversity.

In the technology industry, licensing is a proven model, particularly for companies like ARM Holdings, the British firm and fully-owned subsidiary of Japan's SoftBank that creates microprocessor designs and sells them to architectural licensees.

ARM's fundamental reference architecture processor intellectual property are all part of Apple's A-series, the Qualcomm Snapdragon, Samsung's Exynos, Huawei's Kirin, Marvell's and Broadcom's application processors, and MediaTek's Helio/MT chips. 

And yet, the company doesn't actually manufacture anything. 


Must read


I am not saying Tesla should completely abandon making its own cars. Sure, it should make reference architecture vehicles for boutique customers -- in the same way Microsoft builds the Surface as an outstanding example for what the other OEMs it licenses its Windows 10 operating system to should be building in the mobile PC space. But it shouldn't represent the balance of its business.

Licensed production in the auto and aerospace industry is actually quite common, or at least it was in decades past. Typically you would see this happen in wartime. In Germany, during World War I, Opel received a license from BMW to produce aircraft engines. 

British Aerospace licensed the Canberra design to the US in order to produce its first medium-range jet bomber, and Turkish Aerospace Industries produced a version of the F-16 fighter jet licensed by General Dynamics.

In the past, it was also not uncommon to see developing countries like India, Pakistan, Poland and other companies in Eastern Europe and South America produce Fiat and Volkswagen designs for cars and motors for their domestic automobiles under license. Yes, some of these countries -- such as China -- ended up cloning engines and cars with unlicensed production, and that's certainly a risk Tesla would have to consider under any licensing scenario. 

Also: Tesla's China factory to make 3,000 cars each week

But I believe the benefits far outweigh the risks for Tesla, especially given the significant level of technical complexity and competence needed to replicate the components in order assemble these cars. Plus, the company can act more like a Google in its relationship with Android Google Play licensees, where all source code for the integrated systems would be maintained by it, and could deny non-compliant licensees or unauthorized clones further software upgrades.

Would this represent a major paradigm shift for Tesla? Yes, and to do this Elon Musk needs to swallow his pride, or his board of directors needs to think seriously about removing him from his role, which, frankly, has become more of a liability than an asset as of late. 

But it's early in the company's evolution and it can't afford to act like an Apple when it can't claim the same level of customer loyalty and market penetration. It needs to make some important changes to its business model, and do it now.

Does Tesla need to be more of an ARM than an Apple? Talk Back and Let Me Know.

Who is really in the driver’s seat? Unknown digital threats to your car’s security

Previous and related coverage:

Elon Musk: $35,000 Tesla Model 3 arrives but job cuts coming as sales shift online
Tesla to slash sales force now that the Model 3 electric car is finally ready to order.

Tesla patent aims to boost manufacturing through augmented reality
The new patent could bring augmented reality applications to Tesla factories.

Tesla fatal crash: Parents of dead teen sue over alleged faulty battery
Tesla accused of providing unsafe, fire-prone batteries.

Editorial standards