This week’s podcast is on innovation, which often is discussed in fuzzy terms. I present some of the better language used to discuss this topic. And I present some frameworks from McKinsey and Professor Melissa Schilling (NYU).
The slides mentioned are below.
Related podcasts and articles:
- #32: Innovation, Adaptation and Resilience as Competitive Strategy
Concepts for this class.
- SMILE Marathon: Sustained Innovation
- Dominant Design and Architectural vs. Component Innovation
- S-Curves for Tech Performance vs. Tech Diffusion
- Increasing Returns to Tech Adoption
- Learning Effects and Learning Curve
- Network Effects
- Path Dependency
Companies for this class:
I write, speak and consult about how to win (and not lose) in digital strategy and transformation.
I am the founder of TechMoat Consulting, a boutique consulting firm that helps retailers, brands, and technology companies exploit digital change to grow faster, innovate better and build digital moats. Get in touch here.
My book series Moats and Marathons is one-of-a-kind framework for building and measuring competitive advantages in digital businesses.
Note: This content (articles, podcasts, website info) is not investment advice. The information and opinions from me and any guests may be incorrect. The numbers and information may be wrong. The views expressed may no longer be relevant or accurate. Investing is risky. Do your own research.
Welcome, welcome everybody. My name is Jeff Towson and this is Tech Strategy Podcast. And the topic for today, Intro to Tech Innovation, Elon Musk and Android’s Dominant Design. So that’s kind of a wordy, vague title. I suppose I couldn’t be any less clear than that. But basically what I wanna go into is this idea of technology innovation because it’s kind of been a running theme in a lot of these. talks and competitive dynamics and operational marathon, sustained innovation being key to ant financial strategy. I mean, it’s kind of, it’s been in here a lot of ways and I’ve always sort of been vague about it, partly because it’s a really big subject and I’m not an expert in innovation. I mean, there’s books and classes and expert, I mean, this is a huge subject and there’s a lot of people with deep expertise in it and I’m not one of them. But it does overlap with digital meets strategy, which is my area. So it does come up, and I thought it would be worth going into this in a little more detail, laying out some of the basic big ideas from my perspective. And then talk a little bit about, you know, Elon Musk, Warren Buffett, Android, how they won, stuff like that as examples. And really why it’s such an important factor rate of innovation, rate of sustained innovation, why it’s such an important factor in competition. in the short term and long term. So Elon Musk has famously said on an earnings call a couple years ago that basically moats are lame. Warren Buffett talks about moats, competitive advantage, this is a lot of area I focus on. And he said moats are lame. The only thing that really matters is your rate of innovation. that innovation, your speed in doing so, is the primary determinant of your long-term competitiveness. And the example he cited on that earnings call, which you can find on YouTube or whatever, is he said, you know, you looked at Amazon and you looked at Walmart, and Jeff Bezos was innovating like crazy. And innovation at Walmart was more or less negligible, and it was obvious who was gonna win three to five years in the future. If one company’s releasing two cars per year, and another company’s releasing one car every two years, who do you think’s gonna win? And that’s a pretty compelling argument. Now Warren Buffett has pushed back on this, saying, you know, Elon’s awesome, but he wouldn’t wanna compete with me in candy. Not a lot of innovation in candy, really. And Elon sorta tweeted out, oh, I’m starting a candy company, and it’s gonna be awesome, I’m super, super serious. It’s like literally his quote. So that’s kind of fun, because it’s fun to listen to billionaires argue publicly. But it’s a really compelling argument, or at least it’s a compelling question. But then you step back and, okay, what does that even mean? Everyone talks about innovation. Ask someone to define it. What do you mean by innovation? I can define a competitive mode, a competitive barrier. I can give you an exact definitions, and I can actually measure them. So that’s a pretty good concept. You can talk about gravity. That’s a theoretical idea to describe something in the world, but it’s a very clean, clear definition and you can measure it. We can look at a lot of stuff, but there’s these other ideas floating around like innovation, sharing economy, things like this. They’re just kind of fuzzy. And I find it very hard to, you know, okay, you need to innovate faster at your company. What does that mean? Like if you’re the CEO, would we just do stuff quicker? Like everyone walks faster in the hallway. What does that mean? I didn’t even know what to do. Now I’m gonna try and dig into that a little bit today and sort of put some cleaner definitions on it. But it’s remarkably fuzzy for something that people talk about all the time. I’m gonna cite a couple people here. Most all of this thinking from today is not mine. So I don’t wanna, I don’t wanna. appear to take credit for this. This is going to come from McKinsey. It’s going to come from a woman named Melissa Schilling, who is a professor at NYU, who is sort of an investment or technology management guru. There’s big textbooks on this subject of innovation and how you manage tech companies, and she’s one of the really good thinkers on that. A lot of this is her thinking and layout. And then some of it’s McKinsey. little bit of its mind but now today’s definitely out of my strike zone more than previous stuff. Okay so let’s say we look at tech history. It’s pretty awesome. Like go online and just pull up major inventions in technology, you know, from 1800 today to today. And it’s pretty stunning. Like I’ll read you some of them. Like 1800 to 1820, 200 years ago, what was invented? The electric battery, the steam locomotive, the bicycle, the internal combustion engine and the telegraph. Okay, those seem to be important. Let’s say we go 1860 to 1880. Then someone invented the typewriter, the telephone, the phonograph, the light bulb, also the Gatling gun, which is interesting. 1880 to 1900, light steel that basically enabled skyscrapers, the internal combustion automobile, the pneumatic tire, the electric stove, x-ray machines. 1902, air conditioner, big deal in Thailand. The airplane, vacuum cleaner, electric washing machines, the rocket. I mean, it’s just when you just go decade by decade, it’s amazing, 1928, penicillin, 1927, the television, 1939, atomic fission, 1947, the transistor, 1943, the nuclear reactor, integrated circuit, 1958, 1969, the ARPANET, precursor to the internet. 1971, the microprocessor. 1981, the space shuttle, so on and so on. I mean, it’s really a stunning list and you realize like how much the human race has really changed in the last 200 years versus the arguably 200,000 years of human history prior to that. And then suddenly in 200 years, everything takes off. Well, why? technological innovation, which brings us back to Elon Musk’s argument that that is the primary determinant. That’s the big lever. At a macro level, that’s a pretty good argument. All right. Now McKinsey did some interesting thinking on this a couple of years ago, McKinsey Global Institute, which is a little different than the McKinsey Quarterly and all that. And they came up with a framework for innovation, which I’ve cited before on this class, and I’ve always remembered it. I like stuff when someone tells it to you once and you always remember it. You know, that’s very helpful. And they basically said there’s four types of innovation, four levels, easiest level, most common level, they call cost innovation, which is just making stuff cheaper. So you’re improving your processes in your factory, you’re making things like, I don’t know, tables or cell phones or whatever, and you’re just making them cheaper than they used to be. And that’s not unimportant. It’s not sexy. The people who do this, you know, they’re not on the cover of magazines and whatever, but it’s super important. And that’s gradual. A lot of this happening in China. Chinese factories, Chinese entrepreneurs are very good at making things cheaper. If for no other reason when you do that in China, you expand your market dramatically because there’s several hundred million other people at a lower income level there. So it’s a good way to grow your market. But cost innovation. Okay. Next level up from that, which is a bit harder, customer facing innovation. That’s a bit vague. That’s Steve Jobs. That’s Mark Zuckerberg. That’s kind of half a Silicon Valley. Look, you know, these folks are not creating anything new. Steve Jobs never created anything in terms of, hey, let’s make a new transistor. Hey, let’s make a new type of printer. I mean, he didn’t do any of that. What he did do is he hunted the world for interesting things and he put them together in packages. in products that were very, very compelling to customers, primarily consumers. The iPad, the iPod, the PC, I mean, he was sort of very good at innovating. in bringing things to market, making it usable, making it pleasant, turning ordinary objects into more compelling and powerful objects like a PC or an iPhone, but he didn’t invent any of those components. He wasn’t working in the lab, soldering anything and doing lab experiments. A lot of Silicon Valley is like that, which is why you hear so many stories of, oh, this person’s a college dropout and they started a tech company. Or this person got their MBA and then they started a tech company. That usually means they don’t have a science or hardcore tech background like engineering. And this is what they’re doing. When a company like Theranos, you know, this sort of scandalous medical diagnostics company, when a freshman in college drops out and says she’s going to start a medical diagnostics company, you can’t do that. Like you can drop out of college and do software and create video games because you’re mostly in the customer facing innovation. To do medical diagnostics, you need PhDs in biology and chemistry and you don’t just pick that stuff up. Everyone who works in that field has decades of training because it’s really complicated. You don’t pick up cellular biology on the fly. So that’s why I’m looking back, it was pretty suspect. Okay, next level up, engineering innovation. So we go cost innovation to customer facing innovation. So that’s say, let’s say Chinese factory to Steve jobs, Bill Gates, Mark Zuckerberg. We go up the next level, engineering innovation. That’s Elon Musk in my book. He’s not just putting stuff together. I mean, he’s a serious engineer and physics. you know, professional who was a physics PhD student, graduate student at Stanford. Like those are some of the, literally some of the smartest people that you’ll ever meet in this world, PhD grad students at Stanford or MIT. I mean, they’re off the charts. So when he’s building rockets, he’s not Steve Jobs just ordering people what to do. I want you to do a software that does this, but he doesn’t know how to code. When Elon Musk is building rockets, he actually can build them. He knows all of the physics and engineering and everything. That’s much, much more difficult. So engineering innovation, new types of engines, new types of batteries, new types of materials. You’re much, much harder, much more difficult. Your likelihood of success drops because it’s actually hard to build new machines that work. If you’re building a video game, It may suck, but it’s not gonna fail. You’re not gonna spend two years building a video game and then the thing just doesn’t function. It may not be popular, but it’s gonna function. But you can spend two years trying to build a rocket engine and it’s like this thing just doesn’t work. It keeps blowing up. You’re dealing with the laws of nature at that point. As software is more like the laws that we’ve created in business. Then we move up to the next level, which is science innovation. Biology, chemistry, medicine, physics. This stuff is really hard. There’s only a couple places in the world where you find. people that can do this. Like people say, oh, what countries can do this stuff? Can China do engineering innovation? Sure, because there’s a ton of civil engineers and such. But there’s only a handful of places in the world like Cambridge and Boston where you see science innovation where people can actually spend $500 million doing bench research and come up with a drug that actually does what it’s supposed to do in the human body and doesn’t kill you or hurt you. The rate of failure is… the vast majority. When I was back at Stanford and med school, I hung out with a lot of friends who were MD PhDs. There’s the MD program I was in, but there’s also MD PhDs because this is a huge research school. So I met a lot of these folks were my friends. They were some of the smartest people I’ve ever met. Still to this day, probably the smartest people I’ve ever met. They’re getting a PhD in neuroanatomy while they’re also picking up their MD at the same time. and then they spend all their time in the research labs designing the next generation of MRIs. I mean, it’s staggering. But looking at them now many years later, very few successes. Most of them, what they’re working on didn’t work. That’s the most common outcome. or if they did work it was quite small and a quite limited use case because it’s super hard and most of the stuff you do in science innovation ends up not working. Most drugs don’t work. So that’s super difficult. So those are kind of the four levels. And when I look at so Elon Musk, I think he’s doing engineering innovation, which is difficult but doable. Science innovation, very difficult. Most business people and where I try and live is in the what I would call technical but not technological. Things where there is a technical component to figure out that is tricky, but it’s not based on a new technology. So insurance underwriting, software code, businesses merging with software. I mean, those are hard problems, but they’re doable. That’s kind of where Bill Gates lives. It’s kind of where Warren Buffett lives. Warren Buffett does not try and figure out bench research in science. He looks at underwriting and he tries to figure out a complicated business question with a lot of technical details before anyone else. You know, that’s kind of the, I think it’s kind of the sweet spot. Okay, so that’s a decent framework for this. I’ll put this in the show notes. Cost innovation, customer facing innovation, engineering innovation, and science innovation. Now going forward, I’m gonna sort of narrow this down to my world, which is software meets business. What does it mean to innovate within that boundary, which is really in levels two and three I just mentioned. I don’t go too deep into the tech stack of, you know, how to create new batteries and stuff like that. I don’t, you know, it’s outside of my expertise. So that’s kind of the area I’m going at. For those of you who are subscribers, by the way, this is all gonna go under Learning Goal 32, which is innovation, adaptation. as a competitive strategy. I’m actually going to highlight five or six important ideas today, but they’re all going to go into the heading of innovation. For those of you who aren’t subscribers, if you want to join in, you can go over to jefftausen.com. Sign up there for a 30-day free trial. I would much appreciate that. If you don’t really want to do that, but you’ve been listening, because I know there’s kind of a lot of people that listen to this now, which is awesome, by the way, you can go over to iTunes and give it a review. that would actually be very, very helpful. That stuff does kind of matter. Now there are some sort of standard frameworks for innovation. I’m gonna give you a couple of them here. Just make a mental note that can be helpful. You can kind of define innovation or look at different aspects of it. One common one is product innovation versus process innovation. Okay, that makes sense to you. You come up with a new way of bending metal or making a semiconductor. That would be one type of thing. Or you come up with a new product entirely. And usually those things are actually related. That when someone comes up with a new process for doing something, then you come up with a new product as a result. Process innovation can be, it can be broader than just production technologies. It could be marketing methods. It could even be business models. I would add a third one there, which I would. called business model innovation. And it’s kind of how the organization conducts its business, whereas a product innovation, the output of an organization, that can be a good or it could be a service, could be both. My favorite example of this is actually the ballpoint pen. This is kind of a funny story. A couple of years ago, 2017, there were all these articles in China about how China had finally figured out how to make a ballpoint pen. and I didn’t know it was that difficult. And sure enough, I read into it and it’s like, it turns out making ballpoint pens, that little roller ball at the tip of a pen is really difficult. And for decades, China, like most countries, had been importing the pens, or when they started making the pens, they would import the little steel ball and the little casing around the ball because they couldn’t do it. It turns out the precision machining you do to need to make those little balls is actually, it’s like tiny ball bearing basically. They’ve been importing those from Switzerland for decades. And then you kind of have this high quality steel that goes around the ball and the ink has to go between them. They’ve been importing those from Germany and Japan forever. And it’s kind of an ongoing thing. Like back in 2015, you know, there would be… TV shows with executives about how can China finally crack the ballpoint pen. And like literally the Chinese premier would talk about the fact that like we need to focus on this and you know our pens are too rough and they don’t roll. I mean who knew? But it turns out the process innovation required was quite difficult. And so they finally cracked it. It was all over the news. It doesn’t have to be these massive breakthroughs. A lot of it is just the stuff we see every day. There’s a lot of innovation that required that. So that would be one way to think about it. Product innovation versus process innovation versus business model innovation. You could, another dimension, radical versus incremental innovation. People often call this like disruptive versus sustaining innovation. Some different words here, but the basic idea is like you’re creating something entirely new or You are making incremental gradual improvements in what you have You’ve got a fan. It’s an electric fan every year you make sort of incremental Improvements to the fan the blade how it works the motor all that Versus radical innovation a new type of fan that works an entirely different way Okay, you hear people use different language for this, but it’s usually incremental versus radical or something like that. Competence enhancing versus competence destroying innovation. You know, sometimes you invent the car and it destroys all those companies that had built their competence, their core skills around horse and buggy. You know, it’s very hard to adapt that. You know, when digital photography was invented, that was a competence destroying innovation for Kodak and its ability to make photographs with chemistry. You know, it’s pretty common. So competence enhancing, okay, we’re making this skill set better versus A, we’re just wiping out that skill set. If you’re selling slide rules, horse and buggy, you know, calculators and cars basically wipe those out. Competence enhancing, let’s say each generation of CPUs is a little better than the previous one. Engineers get better at making them, things like that. So that’s more about the skill set. All right, now the big one, which is what I’m gonna talk about for most of this class, is architectural versus component innovation. Sometimes it’s called architectural innovation versus modular innovation. And that word modular, you hear that a lot. And this is the idea that when you start talking about technology in particular things tend to be made of parts a bicycle has certain parts it has wheels, pedals, chain, handlebars, so the architecture is the style of the bike we’re very used to two wheels two pedals, handlebars that’s the architecture and then within that there’s a lot of components so you could have a lot of innovation at the component level new type of handlebars new type of chain or you can have innovation at the architecture level Here’s a new bike where you don’t sit on top of it leaning forward, you sit backwards and it’s got three wheels and who knows what, right? And we see that in things like laptops and PCs and smartphones. You know, there’s a very complicated architecture that has to be put in place. And then you have a system of components. So they kind of say, look, things are nested in a hierarchy architecture. And then you have a system of components. and you can see innovation at both levels. And there tends to be a cycle, which I’ll talk about in a moment, which is tons of innovation at the architecture level until a dominant design is established, like Android’s smartphone operating system. And then once the dominant design, which is an important idea, is established, then all the innovation switches to the component. Okay, put that aside for a minute and let me take a quick diversion then I’m gonna come back to that idea of dominant design. There’s a couple really important so what takeaways, key ideas when it comes to technology innovation. And I’m gonna list for you three to four of them and they’re just important. Like, why is tech innovation so powerful? You know, we can have product level innovation. We can have someone creates new candy bars, someone creates new shoes. Why is innovation and technology so powerful that it just seems to shape the course of industries and to some degree, the human race? Well, not even to some degree, to a massive degree. Here’s why. So what number one, takeaway number one? Technology performance tends to follow an S curve. So I’m gonna put one of these in the show notes, but if you have a, on the X axis, we have cumulative effort, and on the Y axis we have performance. So as we move from left to right, I know you love it when I describe graphs. It’s gonna be in the show notes, but everyone’s seen these, these S curves. As you move from left to right, it starts to slowly increase, and then increases very quickly, goes straight up, and then it flattens out, like a big S. And the Y axis is performance, and the X axis is cumulative effort, which is the more time and energy and money we spend in a technology cumulatively, the performance doesn’t just go up and up, it starts to go up exponentially. It starts to really take off the more time, effort, and attention we focus on a technology, semiconductor. In… combustion engine, any of those things I just said. That gets you that accelerating part of the curve, but then at a certain point, we reach the performance maximum of this particular technology, and then it flat lines out, which gets you the top of the S-curve. Because it’s an engineering aspect or a science aspect, and it will eventually have a theoretical limit. So that gets you the S. Now that’s technology performance. And so these things can have a very powerful effect. And when we see two technologies compete with each other in terms of performance over time, it can actually be kind of confusing. Like we’ll see an S curve for a technology. I’m going to make this one up. Let’s just say an internal combustion engine for a car. And it’s doing the S curve, and then we come up with a new type of engine. Let’s say it’s an electric motor or something. I don’t know anything about these, so this is probably wrong, but just assume it’s true for the purpose. And let’s say when we look at its performance in the early days, it’s actually lower than the internal combustion engine. It would be lower on the y-axis. And we think, oh, this is not so good. and maybe the rate of increase is slower. So the S is not as vertical. But it might be at the end of the day, the maximum performance is actually higher. So it might look like a bad investment. Hey, it doesn’t work as well. Hey, it doesn’t sort of go vertical as quickly. But after enough cumulative effort, we may find that its top tier threshold of performance is higher. than the internal combustion engine is capable of. So these things can be very confusing when you look at them in various stages of development. It could be lower at the beginning, but it could have a steeper curve. So when it does start to take off, it goes vertical faster, and then maybe it ends up at the same place or the little place. You’ve got to kind of think about the beginning level of performance based on the effort you’ve put in. As you put in more effort, how rapidly does it increase? And then where is the ultimate sort of maximum limit for performance based on the science of this technology? So it can be a little bit confusing. An example might be, here’s a better example, since that was not a very good one. Let’s say propellers versus jets. Propellers had much better performance early on and you got a lot more bang in the early days because it’s not that hard to build a propeller. So a certain effort, the planes are flying. Jet engines, much, much less. But obviously over time, jet engines proved to have a performance that propellers are not capable of. And at a certain point, propellers, even though the propellers, the technology never got really to the maximum. It was just kind of abandoned. Once jets came into play, people stopped working on them and the effort shifted somewhere else. So that’s kind of the key idea. So what number one is like tech performance tends to follow an S curve. We don’t see that with product innovation in consumer products and apparel and things like that. So there’s a lot of power there. So what number two, takeaway number two, technology diffusion also exhibits an S curve. And everybody gets this confused with what I just said. When you talk about the S-curve, the first one I was talking about was the performance of the tech based on cumulative effort. This one is more about, okay, the tech exists. Here’s a smartphone, here’s mobile payment, here’s whatever. Here’s a new type of ERP system. Here’s a new type of engine for a vehicle. Here’s a new type of electrical system or a navigation system, whatever it might be, the technology. What is the rate at which it is adopted by the marketplace? At what rate does it diffuse across society or business? And there’s a lot written on this, like the famous book is Crossing the Chasm, that when you start to sell, in that case, a B2B technology, there’s early adopters who are a small group of people, but they use the new stuff. You go to early majority, you go to late majority, you go to laggards. This is either shown as an S-curve of cumulative adoption. or a bell curve is where it’s often shown. I’ll put both examples down below. But people tend to confuse the adoption of technology, the diffusion of technology, which exhibits an S curve with the performance of technology, which also has an S curve. Now, why would a technology, even though it works and it’s growing, why would it be slow to be adopted? Well, there can be some upfront costs. Maybe they’re significant. Maybe there’s some risks. Maybe there’s a lot of uncertainty. As things become more certain and clear and as costs fall over time, then you see the adoption rate tends to take off and you get the S curve. And then eventually it reaches market saturation, which is the top of the S. Okay, so those two things are different. Diffusion is an S curve, performance is an S curve. So what number three is we put both of those ideas together and that gets you Clayton Christiansen’s very famous book, The Innovator’s Dilemma. If you haven’t read this book, read it. It’s awesome. It’ll blow your mind. It’s mostly about B2B technology, mini mills, things like that. But in my opinion, he’s basically combining these previous two ideas, which is, look, performance increase in a technology happens much faster than the adoption of the technology by the market. There’s a disconnect there. the ability of companies or people, the capacity, the willingness to adopt technology tends to be quite a bit slower than the performance increases of the technology itself. And so what you end up seeing is, this is Christiansen’s work, you tend to see companies piling on more and more features that customers don’t really want. Here’s a phone that does all these things. Here’s a camera. Now we’re gonna get advanced lenses. And here’s an ERP system. We’re gonna keep adding features. And the product functionality starts to move way beyond what customers actually want. There’s a gap there. And they’re doing this to compete with each other and to get good margins. And then… The innovator’s dilemma is when you get a low cost rate, often very simple technology will come in underneath that. They call it low end, low, I’m not even gonna try and summarize because I’ll do this badly off the top of my head. But most people will kind of know this idea. If you don’t, look up Clayton Christensen’s work, The Innovator’s Dilemma, it’s awesome. But it’s this idea of customer requirements versus tech improvement. And you’ve often opened yourself up to a simple. low-cost tech, you know, version of your high-tech technology to come in and better serve customers because you’ve moved beyond what they want. Andy Grove, the founder, one of the founders of Intel, he used to call that segment zero. You don’t want to expose yourself to segment zero. And he, I mean, he talked to Clayton Christensen, I think. I think that’s where he got that. Mini Mills did this to the Steel Mills. Anyways, there’s a whole lot there. I don’t wanna get into it because I’m giving a simplistic at best explanation right here, but it’s that main idea. Okay, that brings us back to this dominant design idea. This is a really important tech phenomenon, which is why in the title of this talk, I said Elon Musk and Android’s dominant design. introduction to tech innovation, Elon, and Android’s dominant design. Because when smartphones came out, you had a lot of competition, a lot of innovation, where everyone was sort of fighting at the architecture level. So architectural, architecture innovation versus component level innovation. This was a fight for the architecture of a smartphone. What’s it gonna look like? The operating system. The App Store is mapping important, is mapping not important? Turns out it’s really important. Is Messenger important? There was a lot of fighting to establish the architecture of this new thing called a smartphone. And iOS and Android fought it out for several years. Before the architecture was really finally crystallized and everyone agreed this is what a smartphone is. And then the competition and the innovation shifted to the component level. Let’s make a better camera. Remember when cameras weren’t even in smartphones? You know, let’s make a better messenger, let’s have better payment, let’s have a better app store, let’s integrate the map into Taobao and Amazon and all those things. Everything started to advance at the component level. And I’m putting in a JPEG here, which basically is a circular pattern that you will often see when this stuff happens. You see a technological discontinuity. like the emergence of the smartphones versus the PCs, you see an era of ferment. This is Michelle, Melissa Schilling’s language, not mine, where you have design competition, a lot of substitution where we’re starting to move from PCs, we’re substituting smartphones for that. Then the dominant design gets sort of established and selected, in this case, the Android operating system and iOS to a much less degree. And then it shifts to an era of incremental change. where the dominant design is elaborated, now you see a lot of innovation at the component level. And that, it kind of stays there for a while until another technological discontinuity shows up and it’s a cycle. And most of the time is spent in that second phase where the dominant design is established, there’s a lot of innovation and a lot of specialization by companies and really cultures become wedded to that and companies establish very sort of clear processes within them, the culture, the work flows, all become very established, which is why it’s often so hard for those companies that are successful there to then switch when a new technological discontinuity emerges. I mean, why didn’t Microsoft make the transition to smartphones? They dominated PCs, very good. They had very elaborate systems, very precise processes, a culture that was very focused on that. And then when the next discontinuity came around, they weren’t really well positioned to change quickly. So it’s kind of like winning in that second phase puts you at a disadvantage for the next cycle. Now all of that is Melissa Schilling’s language. So I don’t wanna give the impression that that’s my thinking. She’s a really outstanding sort of thinker on innovation and innovation management, things like that. And that brings us to the final point, which is. Technology innovation has increasing returns to adoption. That’s the point. Technology innovation has increasing returns to its adoption in a business, in society. It gets more powerful as it goes. Now there’s lots of reasons for this. I’ve sort of pointed to some of them like the performance increases, the adoption increases, but there’s really tons of reasons. There’s network effects for a lot of this stuff. Things become more valuable the more people use them. If more people are using PowerPoint, it becomes more useful to other people. More people become trained in it. More people specialize in it. Bigger companies emerge doing that one particular technology and therefore they have the revenue to invest in R&D in that one area. We start to see complementary assets. get developed around a technology. Hey, here’s a railroad, let’s build some towns around that railroad track. But let’s say there are two primary sources for increasing returns to tech adoption and this is Professor Schilling’s argument or statement. The big two, learning effects, network externalities, also called network effects. Now I’ve talked to you before about sort of the rate of learning. as a competitive strength, that it can get you a competitive advantage and it’s also one of the key sort of competitive marathon dimensions in my smile marathon. I was kind of talking about the learning curve, a broader term for that would be learning effects. So there’s actually two concepts here, learning effects versus learning curve. A learning effect is basically the idea. The more a tech is used… the more often it is used, it just tends to become more effective and efficient. I mean, it just happens. It gets revenue. The company’s big money. They reinvested in R&D. Companies get experience using that technology. The staff become better at it. They get more productive when they use the tools. They get better at implementing it in various situations. There’s just a lot of benefits in terms of efficiency and effectiveness the more a technology is used. That’s a learning effect. A sub part of that would be, let’s say, the learning curve. And this is the old, I’ve mentioned this before with the Model T Ford and the more you had cumulative production, the cheaper a factory would get in the old Henry Ford Model T factories. That’s more of an almost a manufacturing idea. As you repeat a process over and over, people learn to produce it more efficiently. You tend to reproduce input costs, wastage, things like that. And you can see that for anything where you’re just doing the same thing over and over and over, making a car, making a McDonald’s or a Subway franchise. The more of them you do, you actually get better at it. So that kind of learning curve is more applicable to a lot of industries, but learning effects is obviously. big thing in technology. A related idea within this learning category would be absorptive capacity. I’m giving you kind of a lot of important definitions here. You don’t have to remember them all, but I wanted you to hear them. As you accumulate knowledge, your ability to learn more knowledge grows. That’s how human beings are. That’s how companies are. That’s how groups of people are. As you learn one part of knowledge, you get better at spotting subtleties in the new parts. You know, we do sort of, you know, if you study one subject over and over, you actually get faster and faster at learning it and you get a more detailed and subtle understanding. And that pretty much works for companies too. So that’s kind of the first big bucket of why does technology become better? Why does it have increasing returns the more it is adopted? Number one, learning effects with a couple of sub points. Number two, network effects, network externalities. I’ve talked a lot about these over the course, so I’m not going to go into this too much. This has been happening for a long time. The old telephone networks of the early 1900s, the more phones that are connected, the more valuable they are to everybody. Railroads, that’s got a network effect. Then we move into the digital world, away from physical networks into connections between people, connections between cars, software connections, marketplaces, audience builders, all of that stuff. We’re in the digital age, so things are connecting more. You get a lot of network effects. OK, so those are kind of the two big reasons. But yeah, increasing returns to technology adoption, it’s a huge deal. OK, that’s most of what I wanted to cover in terms of putting some language around this idea of innovation. But a couple of days ago, I sent out an email to the subscribers about valuing Jingdong. JD, how to think about their business model, but also to think about how would you value this company three to five years from now, and that’s when strategy helps you. I mean, strategy doesn’t help you put a valuation in six months, it plays out over a longer time period. And how do you look at a company like JD, which is a technology company, they’ve got a network and all that sort of stuff. And I was flipping through. this book by Professor Schilling sort of brushing up on this stuff and she had a great framework which I thought was very useful so I thought I’d put that in here which is the idea of the value of technology can be broken into three pieces. You have a technology railroad tracks I don’t know semiconductor operating system whatever it might be. The three levels are the standalone value, and basically the network externality value, which is a combination of your installed user base and your complimentary goods. So let me explain that real quick. But I thought it was actually a nice way to think about this. You know, anything you use, whether it’s a operating system, a railroad ticket, anything like that, right? There’s just a standalone value, you could call that the technology utility value, on its own. Can you sell it to someone? What do they care about? What functions does it perform? What’s its utility? Okay, that’s actually pretty easy to think of. You can think about that like, what was the value of the smartphone, the iPhone, when Steve Jobs first released it? When it was basically a standalone product. You could figure out its standalone value, the utility value. However, then they turned the app store into a platform and it became a much bigger thing. And then suddenly you’ve got the value of the ecosystem. You’ve got the value of the fact that there’s all these apps in the store you can do. It became, you know, there became a lot of what we’d call complimentary goods that became available. So okay, those have a lot of value. We could value those separately, but you might have to think about where the… the product is in its life cycle. In the early stages, you may not see a lot of complimentary goods available that you could place a value on. It might be in the early stage, in the early days of a new product, that most of the value is in its utility value, its standalone value. And then you could also look at the installed base. The network externality value is usually really a combination of the installed base of users and the complimentary goods. If a lot of people are using a technology, it can become more valuable for things like Messenger. if a lot of app developers are making something for the phone, those would be complementary goods. But both of those things may not be there on day one. So you can kind of break it into three levels, standalone value, the value of the install base, and the value of the complementary goods. I thought that was quite an interesting way to think about it. And then you can look at how did Android win versus iOS. You had, you know, five years of fierce competition between iOS and Android to become the smartphone operating system as this emerged. And you could look at it that way. As these two products were competing in year one, in year three, in year five, at each stage, what was the standalone value of the tech, the product? What was the installed base and what complimentary goods were available at what stage? And historically, this is kind of where Steve Jobs lost. I mean, this thing happened 30 years prior with the PC when Microsoft came out and from day one, they really pushed to grow their installed base and the available goods. And Steve Jobs didn’t wanna do that. He wanted it to be a standalone product without any sort of connections or complimentary products and Windows took over and Apple fell. It was kind of the same story over and over, although this time it worked out. Okay, last point. we bring this all back to my world. So okay, there’s innovation. When I look at this subject as someone who does digital meets competition, I mean, this is the Elon Musk point, that innovation is a massive, if not the primary driver of competitive success in a lot of businesses, not in candy bars, but definitely in engines. So, you know, why would that be? Well, I mean… The world’s moving a lot faster than it used to. There’s a lot more technology out there. It’s a lot easier to deploy software than it was to build trains 100 years ago. And this stuff just moves faster. And companies appear to be increasingly relying on revenue from their new products. Like, let’s say we look at a… the R&D spending of various companies, and we say, okay, if you’re spending R&D this year and next year, at what year will that product you’re working on show up in the revenue of the company? And you can map these things out. And what you see in a lot of industries is like, the R&D we spend this year and last year is gonna be 30% of our revenue next year, because products are rising and falling so fast now. And it didn’t used to be that way. It used to be you had a, you know, you come up with a good product, you might get 10 or 20 years out of it. Now it like, it buys you a couple of years, but you better have something else coming up. So innovation helps you grow your revenue. And often cases, it just helps you replace the losing revenue because your current products are becoming obsolete. Even if it doesn’t do that, it protects the margins of your products. because if someone else makes a similar product with a little nicer thing, it might eat away. So one, it’s a revenue mechanism. Two, it’s a protection for your sort of gross profits and things like that. Maybe you should be just doing the same products, but we need to have new processes all the time to keep lowering our costs. Another way to think about it would just be, maybe a company used to may have five or 10 products, but now maybe we all need to have 200 products. Maybe that’s normal. We just keep innovating and we make special products that target each niche. A fact that I wrote down when I was reading her book was Toyota offers 193 car models. Samsung offers 30 different smartphones. Now that wasn’t the case 20 years ago. So there’s the revenue aspect, there’s the competitive aspect, there’s the idea, look, we may just, we just have to make a whole lot of products all the time to increasingly target various niches. with products and product variations and all of that. You know, that’s just the world we live in. Overall, you know, the bar is being raised. Everything’s faster, everything’s more innovative. In certain industries like hardware, engineering, rockets, autonomous vehicles, it’s clearly that’s most all of the ballgame. In other fields, it’s gonna be a smaller factor. In my take on… which dimension you have to sort of run fast every day in. That’s my smile marathon. And you kind of make a judgment call, you know, of what it is it takes to win in your business. But definitely innovation is one of those five. And in certain businesses, it’s everything. In other businesses, it’s probably a minor factor. And in some businesses, maybe it’s not that big of a deal. Okay, that’s it. That was kind of a lot of theory. The main ideas for today, I’ll just read them back to you. I mean, this is all under the banner of innovation, but sort of the key one here is just my smile marathon. Though within the smile marathon, the I stands for sustained innovation. That may be the primary dimension of your business. Other ideas I mentioned, dominant design, I think is really important, which ties to this idea of architectural versus component innovation. I brought up the two S-curves, one for tech performance. One for technology diffusion. I talked about increasing returns to tech adoption. Really increasing returns to tech innovation and adoption. Learning effects versus learning curve, network effects. And all of that goes under learning goal 32, which is innovation, adaptation, and resilience as competitive strategy. This is such a wordy class today. Even I’m getting lost in all the terminology and I really love this stuff. As for me, I’m doing pretty great. I’ve had a great week. I taught my class at Siebes in Shanghai, although I did it remotely, which is always a pleasure. Great group of people, great students. It’s a really well-run school. I mean, it’s just a pleasure to any of the students who are there, if you’re listening in, thank you so much for that. And also was teaching at Sasson Business School at Chulalongkorn here in Bangkok. We started that with a bunch of executives over the last couple of days, kind of a lot of hours of me talking. Three, six, nine hours of lectures in the last two days. When you do these executive MBA things, which I did, that’s how I got my MBA, I did as an executive. It’s great because you’re treated a lot better. You get like food. And when I was doing my executive MBA, they used to FedEx me my books, like wherever I was in the world and assignments. So it’s like going to business school with an expense account. It was really kind of nice. And so yesterday it was pretty great. We had, you know, there’s food and this. The problem is because everyone’s so busy, it becomes a lot more sort of intense. Everyone shows up and you do like six or nine hours on one day, because people can’t come during the week. So there’s a little bit of a trade off there. I did six hours of talking on Saturday and three on Friday. And the students did even more than that. They’re not really students, they’re executives. I think they did nine hours of class on Saturdays, crazy. But it was a lot of fun, great group. And that’s a really, it’s a really well run business school. I mean, I’m really, I don’t know, I guess it’s a bit of luck more than anything else, but I’ve been teaching for over 10 years at various business schools. And this is the best I’ve ever sort of had the opportunity to do. Like both the schools I’m working with are just outstanding and it’s really a pleasure. And that has not been my experience overall. Often there’s a healthy dose of frustration. educational institutions are not terribly well known for being sort of lean and mean in their management ability, generally speaking. But yeah, these two are both really well run, great students. Anyway, so thank you to everyone from Siebs who might be listening in. Thank you to everyone from Sassen who was in my class and will be in there in the next week or two. It’s a real pleasure. I appreciate it. And that is it for me for this week. I’m gonna go play some PlayStation and take it easy a little bit. But thank you so much for listening in. I hope this is helpful and I will talk to you again next week.