Last week some of the ReachNow alumni gathered, online and off, to celebrate the ten-year anniversary BMW’s ReachNow mobility launch. It’s hard to believe it’s been a decade since one of the best jobs I’ve ever had, being part of one of the smartest, most committed teams anyone could wish for. If you read my last post about what to look for in a job, being CEO of BMW ReachNow really did tick all the boxes. I miss those days and what we achieved together.
The reunion got me thinking about an idea I’ve wanted to write for a while: why did the automotive industry swing and miss so badly at innovating in mobility services?
This is part one where we’ll look at the industry overall. In the next post I’ll look at what went right — and what could have gone differently — at BMW’s ReachNow specifically.
Now, when it comes to massive entrenched industries trying to incubate new ventures amid technological upheaval and nimble competition, I have more than a little experience. I’ve been there, done that, and gotten the CES t-shirt more than once. At Harvard Business School, fresh out of Microsoft, I sat in Dr. Clayton Christensen’s class (the author of The Innovator’s Dilemma) and arrogantly argued that his theories wouldn’t apply to software firms. (I was very wrong.)
At RealNetworks and Sony I had a front-row seat to the panic that the internet would destroy the music labels and movie studios’ “plastics distribution business” — their physical media profit center of CDs, DVDs, and Blu-ray distribution. Remember Tower Records? Remember when Best Buy had a physical media section? The music industry tried to fight Napster by building their own download services, only to cede the market to Apple and Spotify. The movie studios tried to embrace streaming while working to maintain their theatrical and home video release windows — creating a series of joint ventures including what became Hulu.
Today all of it has consolidated back into the studios, Amazon bought MGM for the content library, and Netflix is the biggest of them all. The auto industry was just the same thing with a different cast — like a Marvel Zombie Multiverse.
So to give my high-level view of what happened in the game that the automotive companies played and lost, I’m going to use a story I’ve told many times over many cocktails. We’re going to play some poker.
Setting the Table
Let’s think of the mid-2010s automotive companies — the OEMs (original equipment manufacturers such as BMW, VW, Ford, GM, Hyundai, and the rest) — as being seated at metaphorical tables of No-Limit Texas Hold’Em Poker.
Picture the three poker tables where these OEMs and other industry players are seated as the Mobility Table, the Autonomy Table, and the Electrification Table. The companies are playing all three simultaneously — so the investments at one table and the losses at another are always connected. Every bet you make at one table influences and constrains the chips you have left to play at the other tables. Once you sit down, to stay at the table costs money even if you’re not making big bets.
Picture the three poker tables where these OEMs and other industry players are seated as the Mobility Table, the Autonomy Table, and the Electrification Table.
In No Limit Poker, players hold two cards and share community cards on the table. Before each card, a round of betting occurs with no ceiling on how much anyone can wager, which is where the “no limit” comes in. Players with deep stacks can force competitors into bets they’d rather avoid. They aren’t forced to bet; they can fold or check to the next player, but if the next player makes a big bet, a player must match it or go higher to stay in the game. For our purposes, the investments in technology, launching services, and capital expenditures are those bets while those common cards are the shared context like market, technology, customers, and regulations.
Let’s deal some cards.
The Mobility Table
At this table sits every company building services that make it easier for consumers to get anywhere they want to go, on demand, from a smartphone. Who needs to own a car when a ride share, short-term car rental, scooter, or bike is just a click away? Sitting at this table were companies like Uber, Lyft, Lime, Spin, and dozens of others around the world. Like Spotify or Netflix in the media world, they were taking a physical good, transportation, and commoditizing it by making it instantly accessible and cheap. All of these new companies and bets were fueled by enormous piles of venture capital.
The OEMs weren’t going to sit this one out. One thing worth remembering about auto companies is that their customers are not consumers; they’re car dealers who sit between car users and the OEMs. The movie industry was the same for much of its history: studios sold movies to exhibitors (theater chains), not directly to audiences. It’s only in recent years, with Tesla and now Lucid, Rivian, and others, that direct-to-consumer sales have become possible for OEMs in some markets. The OEMs thought they understood how to market to the mass market, but they’d never built a real relationship with consumers in a “we have your credit card, know where you go, and can build a profile of you from gigabytes of metadata” way modern companies love to do. Mobility was genuinely interesting to them for exactly that reason. Every car-share trip was another test drive, another data point, another chance to build brand loyalty for the day that customer graduated from a small urban car to a big, profitable suburban SUV.
So the OEMs pulled up a chair at the Mobility Table — even knowing they weren’t sure how to make money by bypassing the dealer network to sell a service, not a vehicle, directly to consumers. Playing here was going to cost billions over many years.
The Autonomy Table
This was the table that promised an experience that flew directly in the face of decades of automotive marketing. The Autonomous Vehicle table where cars might not even look like cars anymore. No steering wheels. Seats facing inward. A computer-controlled buggy that might be more bus than Bronco.
A lot of tech companies sat down here: Intel, Waymo (Google), HERE, Zoox (bought by Amazon), and others alongside the OEMs. But the loudest player was always Elon and Tesla. Elon has been promising full self-driving (FSD) for years and even his much-hyped Robotaxi launch has been plagued by problems and still relies on human assistance. Elon was the player at this table telling everyone else he just knew what they were holding, that their cards were rubbish, that he had all the aces, and they should fold. He’d keep pushing all his chips into the middle regardless of how bad everyone else knew his hand was. When his stack ran low, he’d fold for a few hands, go launch a rocket, convince the Tesla fan-boys on Reddit that FSD was just around the corner to pump his stock price, then come back to the table and do it again.
All the OEMs knew autonomous driving was going to be critical to their future. It was going to enable, through reduced costs, those Mobility Table services they cared about now. They also knew two other critical things: they weren’t sure how to fit autonomous cars into their world, and it would be years before the technology was truly ready.
If the “ultimate driving machine” had been defined for decades as a visceral feeling of power and connection to the road, what did it mean when the car did all the thinking while the occupant took a call or a nap? At that point, as long as the seat was comfortable and the car was efficient, wouldn’t one brand work the same as another, and just how would a company market that?
Getting the tech right to actually work to the satisfaction of skeptical public and government agencies was also going to be a very expensive challenge. Who is liable when there’s a fatality caused by the AI behind the wheel? Elon used autonomy claims to convince people Tesla was far ahead, regardless of results. The other OEMs were reluctant to take that curve too fast, knowing it was a long, risky investment.
The Electrification Table
Remember “Dieselgate“? The fine folks at VW had a problem: software that detected when a car’s emissions were being tested and then ran the engine more cleanly but less efficiently during those tests. Back on the open road, it would reset to run dirtier, pollute more, but more fuel-efficient. The consumers who thought they were buying a “clean” car were getting a “dirty” car that knew how to cheat the system.
Meanwhile, Tesla burst onto the scene and made EVs a reality mostly by eliminating range anxiety with 250-plus mile range. Governments, eager to reduce carbon emissions, jumped on the EV bandwagon, offering incentives to buyers and manufacturers. Zero-emission vehicle credits let OEMs dip their toes into EV production while still protecting profits with gas-guzzling sedans and SUVs. Pushing EVs and charging infrastructure might even help the public forget about that whole “we promised clean diesel but it was really just cheating” thing.
So the OEMs were eager to sit down at the Electrification Table. The catch? This was the hardest and most expensive table of them all. The Mobility Table meant building a direct-to-consumer services muscle they’d never had. The Autonomy Table meant inventing technologies that had never existed. The Electrification Table meant doing both those things and reinventing the entire automotive supply chain, factory network, dealer relationships, and aftermarket infrastructure at the same time.
The Mach-E’s wiring harness was 70 pounds heavier and 1.6 kilometers longer. We didn’t know what was going on in Tesla engineers’ minds. But now we understand. They had no prejudice. We had prejudice. We’d gone to our supply-chain person and said, “Buy another wiring harness.” Tesla said, “Let’s design the vehicle for the lowest, smallest battery.” Totally different approach. (Jim Farley, Ford CEO)
An EV doesn’t need spark plugs or the thousand other parts in a traditional internal combustion engine (ICE). Regenerative braking means brake systems work completely differently. You’re not building engines and transmissions — you’re building electric motors and battery packs, all controlled by software. Crumple zones have to protect a lithium-ion battery pack from cracking open and catching fire rather than shielding a fuel tank from being pierced. Service intervals for EVs are completely different from ICE vehicles, a real problem when service revenue is a lifeline for the dealer network.
The Electrification Table meant reinventing the entire supply chain, creating completely new technologies, retraining the dealer network, and rebuilding or replacing factories. No OEM could afford to be seen walking away from this table. But getting there was going to cost billions before a single traditional car was replaced.
Calculating the Odds
All the OEMs are now seated, playing simultaneously at three tables. What I’ve been describing is what kept them from winning at each one as much, or as quickly, as they should have.
Real poker is partly about reading the other players: learning their tendencies, watching their faces, developing a feel for how they’ll react under pressure. But the rest is math. There are 52 cards. You know the two in your hand, and as the game progresses you learn the community cards on the table. From there it’s pure probability, using the software between your ears to work the odds given what you’ve seen and what’s been bet.
Which brings me to a key point: software is the key to making Mobility, Autonomy, and Electrification possible. You can’t win at any of these tables just by watching other players. You can’t summon a car share, unlock a scooter, or find a charging station without great software. There’s an old cliche that there were more lines of code in a Mercedes sedan than there are in a Boeing 787 airliner or F22 fighter jet. Recently the CEO of Ford Jim Farley told Car and Driver “Everyone thinks these three things—China, software, and EVs—they’re all the same. No, they’re not the same. The software thing is 10 times bigger to me.”
Autonomous vehicles rely on sensors and cameras controlled entirely by code. EVs can’t connect to a charging network or manage range without software. The supply chain, the customer service system, the data analytics behind it all — software dependent, every bit of it. Unfortunately, for all their engineering and marketing expertise, the automotive OEMs have historically been terrible at software. They were too dependent on vendors, scared of cyber threats, worried about over-the-air (OTA) updates, and thus too willing to accept mediocre software because they believed the real value lived in the hardware or had the wrong mindset about how to create a true “software defined vehicle”. They have only recently tried to fix that through partnerships and joint ventures.
I saw that mindset up close at multiple OEMs — first during my time at INRIX, then at BMW, later as a consultant to other OEM mobility projects. The closest parallel from my non-automotive experience? Sony (yes, Sony!) in the early 2000s — right around the time we were losing to Apple, iTunes, and the iPod. Great hardware (TVs, CD players, mobile phones) but was willing to accept bad software. Make of that what you will.
The Final Table
In the decade since ReachNow launched, most OEMs sold or shuttered their mobility services, scaled back autonomous driving programs to a handful of cities, and are still grinding through an electrification transition that’s well underway everywhere but still lagging in the US. Here’s how it broke down.
The problem was the simultaneous nature of the game. It was simply too expensive for major public automakers to keep funding loss-generating ventures in the short term while VC backed firms were willing to buy the pot by sacrificing profitability.
The Mobility Table was the first one they abandoned. General Motors shuttered Maven, their car-sharing venture, while BMW and Daimler combined their various mobility projects into a series of joint ventures that were later shut down or sold off. A few other OEMs experimented for a while, and I consulted for several of them. Some non-OEM players like AAA bravely soldiered on for a little while, but couldn’t scale car sharing against all the chips and customers that Uber and Lyft had accumulated so early in the game. So the OEMs picked up what was left of their chips and focused on the other tables.
The Autonomy Table is still being played, but it’s a long game. Elon still wakes up on the X office floor, splashes some chips into the middle, and tries to convince everyone he’s holding four aces — but after all this time, fewer people believe him. Autonomous vehicles are really, really, really hard. Dense, old, complicated, constantly-changing cities are a nightmare for sensors and AI to navigate. Waymo and Zoox have Google and Amazon behind them, respectively, so they have the money to keep pushing, learning, and growing. The OEMs are still investing, but in poker terms they’re mostly just paying the blinds, betting the minimum to stay in the game, waiting for the moment to go big.
It was the Electrification Table that consumed all the chips, all the resources, and all the attention, and still does. The capital required to refactor the entire enterprise, manufacturing, suppliers, worker training, dealer networks, charging infrastructure, in-car software systems was staggering. Ford wrote off $19.5 billion in EV investments that didn’t pan out. GM absorbed a $6 billion loss. Don’t doubt that both are still at the table. They can’t leave.
The OEMs, partially thanks to that software problem I mentioned, made inconsistent and range limited EVs for much of the late 2010s. I drive a 2017 BMW i3 today and it’s still a great car for my needs. But back in 2016-2019 it had far too short a range to be practical for the ReachNow fleet, despite our best efforts. It’s only now, in the mid-2020s, that genuinely excellent EVs have arrived from the major US, European, Japanese, and Korean OEMs and that’s despite the rollback of US government incentives promoting them.
The Chinese Whales
What’s keeping all the players at the Electrification Table when they were willing to walk away from Mobility and slow-play Autonomy? They know they can’t leave. It’s an existential game.
The Chinese EV manufacturers are the whales at this table, EV-native players sitting down with huge stacks of chips, backed by scale and state support, perfectly willing to shove all-in on electrification. Their presence changes the whole dynamic: they raise the blinds for everyone, flood markets with capable, aggressively priced EVs, and make it impossible for Western OEMs to quietly slow-roll their transition without losing the pot.
These companies weren’t rebuilding from an ICE infrastructure — they were EV-native from the start. Every Western journalist who reviews a Chinese EV comes away saying the same thing: the major OEMs need to be very scared. Regulations have so far kept Chinese cars largely out of the US, but they’re gaining real traction in Europe. The OEMs can only hide so long before the Chinese whales pull up chairs at the Electrification Table and try to buy that pot too.
Counting the Chips
I’ve been telling this poker table story for years to help people understand why companies like ReachNow don’t exist anymore. There are reasons specific to ReachNow — and I’ll get into those in my next post. But because all these mobility, autonomy, and electrification projects were sponsored by and deeply connected to the major automotive manufacturers, events far beyond any one company created ripple effects everywhere. The tables were always connected.
You can’t count your chips until you’re at the final table. And definitely not until the last card is dealt.
(Coming Soon) Part 2: What went right and wrong at ReachNow →
