The March of Folly in Tech

[originally posted on June 18th 2021]

From Uber/Lyft's ambitions to build a self driving car to Verizon/AT&T in internet services to "everyone" building a phone - the tech world is full of folly's, predetermined failures

Barbara Tuchman, in her excellent book "The March of Folly: From Troy to Vietnam" defines folly as:

"Pursuit of Policy Contrary to Self-Interest." In THE MARCH OF FOLLY, Tuchman examines 4 conflicts: The Trojan Horse, The Protestant Secession, The American Revolution, and The American War in Vietnam. In each example an alternative course of action was available, the actions were endorsed by a group, not just an individual leader, and the actions were perceived as counter productive in their own time. Many individuals are guilty of folly (Tuchman also calls this woodenheadedness), but when governments persist in folly, their actions can adversely affect thousands, even millions of lives. Folly is a child of power. "The power to command frequently causes failure to think." From the Publisher

In a nutshell - governments making decisions which were obviously doomed to failure from the beginning but were pursued regardless of self interest or data. In tech we see many of follys as well, driven by extremely profitable companies (i.e. means) and smart executives fearing being disrupted (i.e. motivation) the way they disrupted their competitors in their creation. Fear and means lead to doomed adventures in "other companies backyards" with no skills or theory on why said company will be successful there.

This week showed us several examples. Verizon is selling Yahoo and AOL, a move that was obviously a mistake from day one. In general, Telecoms globally have burned Billions of dollars trying to build consumer internet brands. This glaring lack of awareness is a great example of Tech Folly. What makes monopolies, whose expertise is managing complex networks and selling them to a captive audience, think they can be successful in the hyper competitive and fast moving internet services space? Yet they continue to do this again and again under the guise of "moving up the value chain" as defined by some overpaid McKinsey consultant in powerpoint? Imagine a self aware telecom understanding that their expertise is running networks, embracing being a "dumb pipe" and focuses only on speed, reliability and price? Imagine if a PE firm bought Verizon and modernized the organization, shedding all the "value add services", tripling the speed of the network and cutting prices in half. They would rule the networks AND gain customer love.

Another great example is Phones - Apple and iOS is a threat to everyone in the technology eco-system, that is obvious, but is the solution really to go out and build a competing phone? Why would any software company think they can do what Samsung struggles with, i.e. take on Apple with its design expertise, supply chain, brand and decades of experience building sophisticated hardware? But yet, Amazon, Facebook, Microsoft all tried and failed. Google is still trying...

A similar adventure ended this week as Lyft sold off its self driving team following Uber selling its team as the Ridesharing companies finished shedding their self driving ambitions. This is a space I know well and I had these conversations with the founders of the different Ridesharing companies many years ago. To me, this adventure made no sense - self driving is the biggest technical challenge of our generation, rivaling going to the moon, the atom bomb or the internet but, unlike those, it is being funded by private companies and not the government. This challenge requires billions of dollars of investment with NO revenue potential for decades, since RoboTaxies require full level 5 self driving or have nothing (very different from car companies like Tesla that can sell 80% self driving as a feature, complimenting their core product, a car). Alphabet, one of the most profitable companies ever and one of the most sophisticated technology companies with a long history of taking on, and solving, deep technology challenges, is still struggling to crack this. So several new age Taxi app companies will be the ones to make this happen?

I raised these questions to the founders of the Ridesharing companies and got similar answers to the tech companies building Phones - we might get locked out of the ecosystem, it is business critical, a strategic threat etc. These are all motivations to do something but not answers to the question "why do you think YOU will be able/good at it". Many of these tech follys begin as "strategic risks and mitigations" - not as anything unique - no product/business/marketing innovation or insight. Here are all the reasons why, years ago, it was obvious that they would fail:

1. What technical assets do Ridesharing companies have in developing deep research and technology? What hard technical challenges have they ever addressed before going after the hardest one?
2. What business model will support this? They are not profitable, low margin businesses - how will they fund a multi decade engineering adventure with an all or nothing proposition? How will they win the talent war was needed for this?
3. Why do they think that having a small self driving team will be leverage against technical and financial powerhouses like Alphabet (or whoever cracks the self driving challenge first)?
4. Why is their initial premise that this will be a winner take all market (thus they must be in it) but yet believe that they can be that winner?

These were obvious issues at the start, not some backward looking rewriting of history. I would argue that few executives at these companies believed they would be successful but did not have the courage to walk away. I remember warning one founder that "research is addictive, you will achieve 80% really fast and it will look really cool and get everyone excited but this means nothing in the quest of achieving 100% L5 product. You will sell the concept to your board, investors, the press and will not be able to back out of it even when it's obvious you can't win since your valuation will be tied to it". I am not sure who is at fault but the board of directors who approved burning billions of other people's money to pursue a folly such as this should definitely look in the mirror and ask some questions.

What could they have done differently? Well, at the core, all strategic threats need to be looked at in a SWAT type way, with extra focus on YOUR weaknesses and strengths. Here is what I proposed:

1. Due to regulation and production challenges, it is more rational to believe that several companies will reach the finish line with very similar technology and performance. In this world, those who own the users and brand will be the winners. Does a Rider really care if a human or a machine is driving them? It's about price, convenience, brand etc. In the future world of self driving tech YOU will be the largest customer and will have all the leverage.
2. Focus on your core business - acquiring users, retaining them, building a brand, delivering a ride in less than 3 minutes and when someone has L5 self driving service somewhere, treat them as another driver. They will not be able to build a consumer facing service with full coverage fast enough to compete. Believe in yourself, it was very hard for you and will be harder for someone in the future
3. Be self aware - you are not a "tech" company (look at your margins) but a taxi company. First Ridesharing company that understands that, moves their headquarters out of San Francisco to Atlanta, focuses on its margins, realizes that they do not have the tech leverage that pure play software companies have will win. WeWork could have been a great real estate business if it stopped thinking it was Google (just without the tech and margins). Telecoms can be phenomenal companies if they stop trying to be Facebook and invest in their core business.
4. Leverage your strengths - every self driving company needs data, beta sites, testing environments. You have it. Don't compete with them, partner with them. Build the skills in how to manage a mixed use network (humans and machines driving), how to service these cars, be the largest customer in the self driving space instead of a tiny vendor.

I believe strongly that Ridesharing companies need to be honest with themselves and build a healthy, low margin but large scale business and one day, in the far future, improve their margins with self driving. The most powerful companies in the tech world (outside of China) have basically one product whos market keeps expanding - Google (Ad supported search), Apple (consumer tech hardware), Amazon (selling and delivering physical products), AWS (selling cloud tech), Microsoft (Enterprise Applications), Facebook (Ad supported social networks) - there is so much to do in mobility, in your core expertise, in normalizing your cost to your margins, in merging different mobility methods - keep winning in what you are good at.

There are many such follies going on today for the same reasons and will probably have the same results. Facebook's Oculus, Apple (and all platforms) media business, any non-network service offered by any telecom, Googles phones.

Since Follys seem doomed to happen, the bigger challenge is how to quickly identify them and kill them. Amazon is a great example of how it killed its phone ambitions quickly and ruthlessly, Microsoft's killing of many of its follys when Satya took over. The best move is not to launch a folly, the second best is to kill it quickly, the worst is to let it drag on.