Thrust and lube

2 Comments

Get the short URL

I’m presenting a somewhat ambitious session at Startupfest this week. This is a rough draft of my speakers’ notes, which I’ll come back and annotate sometime later. The speech has changed a bunch based on feedback since I wrote this, but hopefully there are some nuggets of useful in here.

You were built to think

We’ve been fighting friction since we first swung down from the trees.

Humans are different from other species largely because of our use of symbols—our ability to think representationally, to speculate, to play “what if?”, and to communicate abstract concepts through language.

We have huge brains. The human brain is only 2.5 percent of your body mass, but it consumes 20 percent of the food you eat and the air you breathe. Our gigantic brains come at a price. In the first months of birth, a baby’s brain consumes nearly all of the calories it ingests. 85% of brain growth happens in the first three years.  It has to grow quickly to reach adult size; it was small when born so that the head didn’t kill its mother on the way out.

Humans have a higher brain-to-body weight ratio than any animals except birds, and that’s because birds have to be vanishingly light in order to fly. In other words, you were literally built to think.

But your brain writes checks your body can’t cash.

The real world crushes your dreams

The real world undermines our lofty aspirations. Atoms get in the way. We’ve dreamed of pyramids, plumbing the depths of the abyss, and flying through the air, only to realize that moving rocks around, holding our breath, and staying aloft is hard work.

Giving all that organic attention to the brain leaves the rest of the body wanting. We’re not that strong; we have small lungs; we can’t fly. We don’t even have very good senses. Just about the only thing we’re best at is long-distance running.

Because we’ve got good ideas and lousy containers, much of what we’ve been inventing for the last million or so years has been tools. The lever, the wheel, and the pulley gave us leverage over the physical world. They mitigated the limitations of our containers. A human might be weaker than a horse, but given a big enough block and tackle, the human wins every time.

The printing press is another example of this. We don’t have a bee-like hive mind; but we can print books, broadcast programming, and teach the world.

Representative government is yet another hack to overcome the coefficients of friction. We can’t send everyone to Washington, or Ottawa, or London; so we elect representatives to reflect our intentions by proxy (at least, that’s how it was supposed to work.)

Fight thrust with thrust: the industrial era

We’ve been trying to get better at pushing atoms around since we crawled out of the primordial ooze. That’s why the industrial revolution was so amazing: for the first time, machines promised spare muscle, a beast of burden that never complained. Steam would set us free.

We saw an explosion of innovation, commerce, and ingenuity on the backs of those machines. We thought we’d found the answer: fight friction with friction.

In the old world of machines, economists and business strategists tell us that scale is key. If you control the supply chain, the channel, or the resources, you’ll win. If you have patents and costly infrastructure others can’t copy, you’ll win. If you get big you’ll win. Traditional, thrust-thinking business strategy says go after cost leadership and get large fast. Economies of scale and network effects are your friends.

But there’s a problem with big. As organizations grow, they have to coordinate, and that coordination becomes overhead.

Hierarchy and military structure

In the mid-nineteenth century, it was clear to Europe’s military leaders that a single commander couldn’t direct a war fought with many fronts and massive armies. The Prussians concluded that an army whose members were interchangeable and could act autonomously would work better in a distributed battle. The arrival of telecommunications eventually let commanders issue order to troops more easily, but the idea of delegating tactical decisions to trained officers in the field, remained.

While formal structures had existed for a long time in the military prior to this, it took the emergence of large organizations and governments to create the need for management in business and government.[1]

In the 1850s, railroads were growing rapidly. Rail systems were huge undertakings, requiring substantial up-front investment and the creation of corporations that would run them. Unfortunately for their investors, small railroads turned a profit, but large ones didn’t.

Daniel C. McCallum, general superintendent of a large railroad line, observed a manager of a small railroad fifty miles long can be “constantly on the line engaged in the direction of its details; each person is personally known to him, and all questions in relation to its business are at once presented and acted upon” but that a railroad five hundred miles long is entirely different because “any system which might be applicable to the business and extent of a short road would be found entirely inadequate to the wants of a long one.” [2]

McCallum concluded that a hierarchy of management was the solution. Dividing his railroad into smaller sections, he placed a superintendent in charge of each section and gave them specific tasks, such as reporting standardized information back up the chain of command. The result for McCallum’s line—and those of other large railroads who copied this approach—was that large ones could thrive. The transcontinental rail systems were born, and dominated industry.

The industrial era, and the formation of the corporation as a business entity, gave rise to large organizations in other industries. Many of these industries grew vertically, owning an entire supply chain from the resource to the buyer. As they grew, they borrowed from McCallum’s approach, but rather than dividing the organization by geographic boundaries, they chose to organize themselves by functional group. Even governments—who until that time had been focused on land ownership and the maintenance of a standing army—adopted hierarchies and forms of representation to deal with geographic and functional sprawl.

The management models of the industrial era are with us today. They were created to deal with the two problems we’ve seen earlier: A lack of awareness about what’s going on elsewhere in the organization, and an inability to get context on a task quickly. By providing clear structure, roles, and responsibilities, a large organization eliminated the need for awareness—one worker didn’t need to know what others were doing in order to add value to the organization—and reduced the importance of context—since workers were replaceable components in the company as a whole.

In those organizations, only knowledge work varied. Groups of knowledge workers—such as an executive team—could deal with changing strategies and adapt to new problems because the team was small enough that they were within the Dunbar Limit—the 150 or so people that the typical human brain can keep track of.

In other words, the top of the pyramid was allowed to think.

Today, however, knowledge work is the core business of the organization. That same adaptability and strategic thinking is essential for other jobs within the organization, particularly within businesses like financial services, consulting, distributed franchises, vertically integrated industries, architecture, engineering, and information technology.

This is the problem that banks, utilities, and carriers face: there are inarguable advantages to size (they’re called network effects for a reason) but the cost of being big and functional is very high, and the solution is usually to impose structure, procedures, and processes, to turn the organization into a machine, and treat employees like lines of code.

The hole in the middle problem

Being big is still a path to success—that’s why corporations tend to grow and acquire. There’s good research to show that companies with a high market share are profitable, but so are those with a low market share. It’s those in the middle that struggle.

Marketing strategists call this the Hole In The Middle problem. Big companies focus on cost leadership (think Wal-Mart) and small ones focus on a profitable niche (think your local gluten-free cupcake shop.) It’s the guys in the middle that are less profitable, because they haven’t got a good competitive advantage.[3] They’re too big to focus, and too small to compete on cost.

Killing new ideas and squeezing margins

Big companies tend to do two things: eliminate innovation, and compete on cost. Both have bad consequences.

Consider a big company like a phone carrier can only function with militaristic, standardized processes and procedures. That means it not only fails to innovate, but it also infects its economic partners with risk aversion. Consider Steve Jobs’ assault on phone carriers. He had to fight AT&T on innovating around features like voicemail; the carriers expected to dictate phone standards to him instead.

It’s even worse. The carriers now have gained the upper hand in terms of the power of the relationship with the handset manufacturers. And they’re starting to tell the handset manufacturers what to build. And if Nokia and Motorola don’t listen to them, well, Samsung and LG will. So the handset manufacturers are really getting these big thick books from the carriers, telling them “here’s what your phone’s gonna be.”

The second consequence of a big company is that an unchecked race to the bottom on cost is a bad thing for all involved. The following three diagrams show how a race to compete on costs without innovating on productivity usually plays out.[4]

Such organizations survive despite this because they find a way to benefit from scale that’s not based on competition.

  • That may be brand recognition (Coca-Cola, for example.) Humans love familiarity, and when faced with novel choices, the familiar wins. That’s why the Box Office and the App Store are packed with the names of board games, cartoon characters, and magazines that we know well from bygone years: they capitalize on our affinity for the past.
  • Perhaps it’s lobbying for favorable trade (as many large industries and unions do) to get subsidies, import restrictions, and other external advantages.
  • Sometimes, they become “too big to fail” (carriers, automakers, banks, airlines) and enjoy protections and cash infusions from government.
  • Often, they turn into a utility, with legislative or practical barriers to entry (there’s only one strand of copper to a house; which was the result of a state-sanctioned monopoly.)

Whatever the case, once large organizations stop competing on niche and start focusing on cost leadership, it’s a long, slow spiral to the bottom.

But like I said, that’s traditional “more thrusting” thinking. Thrust pushes harder; lube makes it easier. We wanted more lube, and less thrusting.

The efficiency paradox

The answer wasn’t to fix friction with friction. No, the answer was to dramatically reduce the coefficients of friction themselves.

The future isn’t thrust, it’s lube.

In 1712, a brilliant engineer named Newcomen designed a steam engine that would pump water out of mines using the power of steam, condensation, and vacuums.

The Newcomen Atmospheric Engine engine kicked off the entire industrial revolution, both by making coal mining safer, and by making steam engines more than just novelties. But it wasn’t very efficient.

Between 1763 and 1775, James Watt improved on the design. His steam engine was dramatically more efficient, consuming 75% less coal to do the same work.

In his 1865 book The Coal Question, English economist William Stanley Jevons presented a paradox. He observed that even though Watts’ engine was more efficient, and used less coal, the consumption of coal dramatically increased after it was introduced.

Shouldn’t a more efficient engine reduce consumption instead of increase it? Jevons concluded that an increase in the efficiency with which a resource is consumed tends to increase (rather than decrease) the use of that resource.

We see this all around us. Think about Moore’s Law: computers double in performance every eighteen months. That means a server today does the work of two servers a couple of years ago; and four a couple of years before that; and then eight, and so on.

Follow this to its conclusion. A server today is a rack of computers six years ago. It’s a row of racks twelve years ago. It’s a data center eighteen years ago.

Think about that for a second. Today’s server is a Data Center in 1994, when we were just launching the web.

So Jevons would ask: why are we still building data centers? The answer, of course, is that efficiency drives consumption. The number of things we’ve figured out how to do with computers has outpaced their ability to handle that demand. Bill Gates’ vision of a computer in every home had some unintended consequences: today, it’s a computer in every pocket, doing far more than he ever anticipated.

Abundance, not scarcity

Mark Andreesen, who knows a thing or two about startups, said that “software eats everything.” I interpret this to mean that lube trumps thrusting (with apologies to Mark, though I rather think he’d like that.) When an industry has digital inputs and digital outputs, everything in the middle becomes software—and those who fail to catch up, die.

Today, every edifice we’ve erected to fight friction is crumbling in the face of a frictionless future. Entire industries that depended on data are vanishing, because the barriers to entry they’d relied on are irrelevant—or worse, boat anchors stopping them from changing. Blockbuster and Barnes & Noble’s network of stores are obvious examples, but there are plenty of others, such as travel agencies. And we’re just getting started.

When friction goes away, things that were scarce become abundant. When something goes from atoms to bits, there’s plenty to go around. Take what Cory Doctorow calls the Internet’s Original Sin—music piracy. It’s hard to steal a CD, because you can’t copy atoms. But it’s trivial to steal a file, without any loss.

In 1995, Doc Searls published a visionary piece of writing called The Web And The New Reality. He talked of a polyopoly, a world of abundance where sharing was the default. Utopian? Definitely. But he was on to something fundamental. JP Rangarswami explains on his blog[5]:

When you deal with things that are scarce, you can make money with those things. Scarcity has value. If you have the only Penny Blue you can command your price. But scarcity is not the only way to make money. You can make money out of abundance as well.

Linux is abundant. IBM used to make money with its operating systems. Then, when Linux came along, IBM saw a way of making money because of Linux rather than with it. There are many bloggers I know who don’t charge for their blog or allow any advertising. They don’t make any money with their blog. But they still make money because of their blogs.

Examples of this are everywhere. Jeff Jarvis talks about the difference between an abundance and a scarcity mindset in the publishing world.[6]

[Publishers] think content is that which we make because we are content people—we see content as a scarcity we produce and control. Facebook and Google, on the other hand, see content everywhere—in the allegedly useless creations, chatter and links made by people in the course of their lives. They see content as an abundant resource to learn from, value and exploit.

The problem is, the media is not built for relationships because [it] was born in a time of factories, not services.

New businesses make money from abundance. They don’t think like steam-powered factories, fighting thrust with thrust. They think like a coral reef, seining value and nourishment from the waters in which they grow. You don’t charge to be the reef—you thrive because you’re the reef.

Fighting thrust with thrust is so Victorian. But thrust and lube—that’s sexy.

Fire, aim, ready

But wait! Even more has changed in a frictionless world. The old barriers to entry are down. It’s easy to start something. Instead of infrastructure, use clouds. Instead of marketing, use social media. Broadcast? Use Youtube. Fundraising? Kickstarter or Indiegogo. Employees? Mechanical Turk. Payment? Paypal. The list goes on and on. Most of the headaches of the small business have turned into well-oiled services with little or no upfront costs.

Old business models were ready-aim-fire. First you built a business plan; then you got funding to act on it; then you launched. But today’s Lean Startup puts that in reverse. Aim-fire-ready is the new model. Fire wildly in all directions until something goes “ow.” Then keep firing in that direction until you can hit something reliably. And finally, go after that thing with maniacal focus. In other words, prove a market, then a product, and then a business. Why fire-aim-ready? Because in a frictionless world, the bullets are free.

Analyze all the things

There’s yet another side effect of all this frictionless interaction. Because it’s digital, and because it’s trivially easy to instrument any kind of activity, everything is an experiment. Imagine you’re a brick-and-mortar retailer: putting RFID tags on everything, or tracking every customer in your store, is going to cost a bunch of money. With a digital world, tracking is built in. Every action leaves a trail of breadcrumbs, Big Data to be mined.

As connectivity pervades our lives more and more, through Fitbits and mobile phones and Wifi scales and car GPS and myriad other tools for tracking our lives, everything is up for analysis. Everything can be optimized, evolved—in minutes rather than millennia.

Embrace your inner cyborg

So how do we make sense of all this data? The good news is, as we’ve done in the past, we’re building tools to do it—thanks to reduced friction. Take Reddit, for example. It’s a perfect amalgam of humans and machines that finds the best content on the Internet. Just because it doesn’t look like Arnold Schwarzenegger doesn’t mean it’s not a cyborg. There are human-machine blends everywhere: Google Maps, on mobile devices, learns from drivers what congestion is like. What’s better than humans or machines is humans and machines.

There’s a reason for this. Machines and data are very good at optimizing for what are called local maxima. But they’re lousy at inspiration. In his book A River Out Of Eden, Richard Dawkins uses the analogy of a flowing river to describe evolution. Evolution, he explains, can create the eye. In fact, it can create dozens of versions of it, for wasps, octopods, humans, eagles, and whales. What it can’t do well is go backwards: once you have an eye that’s useful, slight mutations don’t usually yield improvements. Put another way, a human won’t evolve an eagle’s eye, because the intermediate steps are all bad eyesight.

Machine-only optimization suffers from similar limitations as evolution. Imagine I gave you three wheels and asked you to evolve the best, most stable vehicle. After many iterations of pitting different patterns against one another, your evolutionary algorithm would come up with a tricycle-like configuration. But it wouldn’t say, “you know what? Four wheels would be way better!” Math is good at optimization; humans are good at inspiration.

Some concrete examples

The vanishingly small costs of a frictionless world are changing many other things, too. I’ll end with three examples to think about: taxis, classrooms, and supply chains—three things that don’t sound very dot-com. Yet they’re all in upheaval because friction has changed the world, flipped it around.

Taxis and Uber

The New York taxi industry had a good thing going. It controlled the dispatchers, which meant that it controlled the taxi fleets. A New York taxi medallion goes for around half a million dollars, according to Medallion Financial Corporation, which lends money to drivers. There are 13,087 taxis in the city with medallions. That’s $6.5 Billion dollars.

What does that medallion get you? Basically the right to pick someone up without planning it ahead of time. Lube to the rescue—with Uber, passengers and drivers can connect for a trivially small cost. The barrier to entry is gone, and a multi-billion dollar industry is disrupted. Better: Uber can use all that data to tell drivers where to hang out to have the greatest likelihood of a pickup.

Flipping the classroom

Let’s talk about classrooms. In a high-friction world, teaching is a distribution problem: how do we get the learning out to the edge. It’s about finding the best teacher in the village. But in a frictionless world, we don’t have to settle for the best teacher in the village. We can have Neil Degrasse Tyson teach us about the moons of Jupiter for a day. Sites like the Kahn Academy, and universities like MIT and Stanford, put content online for free. The classes don’t cost anything today. In a recent movement known as flipping the classroom, kids study content at home, and work on problems at school. They get better grades, too. Teachers have gone from broadcasting knowledge to making it stick because friction is gone.

3D printing

Finally, consider supply chains. What happens when the 3D printer is everywhere? Sure, today there are plenty of skeptics, and 3D printing is the stuff of neckbeard hipsters with solder burns on their fingers. But that’s where personal computers were a few years ago, too. Once, we went to Kinkos to print color copies; now, manufacturers give away the printer and make it back in ink. What happens when every home has a printer? Can car manufacturers simply put a cheap lathe in the service station, rather than stocking spare parts for their cars for a decade? And how will Fedex feel when it stops delivering toys, and starts delivering the raw materials needed to print them?

If I were launching a company today—a billion dollar company—I’d spend all my time figuring out how an industry deals with its biggest point of friction, and then thinking about what happens when that friction vanishes. Solve that, and the industry won’t know what to do; you’ll have flipped it around completely. They’ll bring thrust to a lube game.

And that, right there, is why there’s simply no better time in human history to be a disruptor. The industrial age was fun, because it gave us thrust to fight the coefficient of friction of atoms. But the connected age is magical, because it eliminates friction altogether. Today’s world lets you build anything you can imagine, if you can just:

  • Embrace abundance and make money from because
  • Use free bullets to fire first and scale later
  • Be data-driven and act like everything is an experiment
  • Find out how an industry overcomes friction, then build for a day when the friction is gone.

Enjoy the lube.

 

 

 

 



[1]           See http://www.analytictech.com/mb021/orgage.htm, which references “The Organizational Age” by Rodney Stark in Sociology, 3rd Edition

[2]           Chandler, 1962

[5]           http://confusedofcalcutta.com/2012/04/17/three-little-words/

[6]           http://www.guardian.co.uk/media-network/media-network-blog/2012/feb/15/what-media-learn-facebook