Tuck Knowledge in Practice Podcast: A New Strategy for the $75 Trillion Industrial Economy

[This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of the Tuck Knowledge in Practice Podcast is the audio record.]

Vijay Govindarajan: “Spotify has a music graph. Google has a search graph. Netflix has a movie graph. Because the movie studios only knew how many movies they produced, perhaps the box office collection. But a movie studio did not know what movie VG was watching in real-time. Netflix knows the moment you know product as consumed as opposed to product as used, strategy changes.”

[Podcast introduction and music]

Kirk Kardashian: Hey, this is Kirk Kardashian and you’re listening to Knowledge in Practice, a podcast from the Tuck School of Business at Dartmouth. In this podcast, we talk with Tuck professors about their research and teaching and the story behind their curiosity. Today, my guest is Vijay Govindarajan, affectionately known around Tuck as VG. VG is the Coxe Distinguished Professor at Tuck and a world-renowned expert on strategy and innovation. He is a New York Times and Wall Street Journal best-selling author, and a two-time winner of the prestigious McKinsey Award for the best article published in Harvard Business Review. Earlier this year, VG published a new book titled Fusion Strategy, which takes a deep dive into how AI and real-time data will transform the $75 trillion industrial economy. In our conversation, VG talks about the genesis of his fusion idea and book and the potential it has to reshape how giant industrial firms like John Deere and General Motors design products. You’ve been teaching and writing about strategy for a long time, and advising about strategy for a long time. When did you sense that a new approach to strategy was going to be necessary for the industrial sector, that kind of led to this book?

Vijay Govindarajan: It’s a great question, Kirk. I want to perhaps go back and put this book in context. I’ve always been interested in innovation. That’s my core topic that I’m interested in. What happened was in the early 80s, I started my academic career at Harvard Business School. And at that time, they were asking faculty to do research on the following question. And the question was, how would information technology, we didn’t call it digital at the time, how would information technology affect the world of business and the discipline of management? Here we are. The year is 2024. We are asking the same question. The biggest difference in 2024 is digital is affecting individual human beings—the way we live, the way we work, the way we transact, the way we listen to music. We work from home now. We transact using digital currency. We listen to music in iPod and now iPhone. Now, why is this important? This is important because this huge change on how it is affecting individuals started with the introduction of this device that I’m holding in my hand, which is the iPhone. I took a look at the market cap, the top ten market-capitalized companies in the world in January of 2007.

[Media Clips]: Today, Apple is going to reinvent the phone. And here it is.

VG: And the reason I chose the date was iPhone was not introduced.

[Media Clips]: Wow.

VG: Of the top companies, there was only one tech company.

[Media Clips]: Wow.

VG: Microsoft.

[Media Clips]: Every so often you experience something so new. So delightfully unexpected that there’s only one word for it. Wow. Introducing Windows Vista.

VG: Everybody else was a physical asset, heavy physical asset company. Manufacturing powerhouses like Toyota or General Electric.

[Media Clips]: Water desalination from GE.

VG: You do the same analysis today. The top ten market capitalized companies are all digital giants. The Apple and the Amazon and the Google. The digital giants that you see today are different than the information technology giants we saw in the 80s and 90s.

[Media Clips]: It’s not the printer, it’s the computers. Hadley downloaded a virus off the web.

VG: This is not IBM.

[Media Clips]: Hadley. IBM protects your network.

VG: This is not Dell.

[Media Clips]: My roommate told me that I could get Dell on the internet.

VG: This is not Hewlett-Packard. Because those information technology giants in the 80s and 90s passively delivered products. They simply automated processes of the current business and current strategy. What the digital giants have done in the last 17 years is they have changed strategy. Google changed strategy in advertising. Netflix disrupted the way we watch movies. Airbnb created a new business model in travel. The really important point here is it is only in the last 17 years we have seen how digital has been used to create new strategies, and that happened only in the consumer sector. The consumer sector is only 25 percent of the world’s GDP today. The world’s GDP is $100 trillion, of which only $25 trillion is in the consumer sector.

That is all that has been digitized so far. One sector with one technology, the smartphone. What has to be digitized is the industrial sector, which is $75 trillion. And the important difference is in the consumer sector, we either dealt with pure information goods like Google or analog products which became 100% digital, like cameras. Whereas in the industrial sector, the physical product will never disappear. A car will not disappear. A tractor will not disappear. Aircraft engines will not disappear. But the value will migrate from physical assets to insights using data and AI. That is what the Fusion Strategy book is all about. It is about how industrial companies can actually do breakthrough innovation. In this instance, how they can create new value, not by making their current hardware more reliable but by leveraging data and AI to get more insights from your current hardware. That’s what the book is about. And the price is $75 trillion. That’s why it’s a very important question.

Kirk: That isn’t the price of the book, though, right?

VG: No, it’s $35, but $35 will tell people how they can make trillions of dollars.

Kirk: Well, let’s talk about a concrete example of fusion happening today. In the book you talk about John Deere a lot, but you could talk about another one if you want. But give us an example of fusion happening today.

VG: The best way we can think about fusion is if you think about any analog products and ask yourself the question, what is the value that is trapped in that analog product which can actually be unlocked by digital technologies? Take, for instance, a piece of glass. We have glass buildings. Think about the John Hancock Building which is full of glass. Think of airports. There are many airports which are just glass structures. That is an analog product, an analog glass structure. Many office buildings, in fact, I was recently in one which was 100% glass, and I was making a presentation to the CEO and the top management team. Immediately they brought the blinds down so that my presentation would be visible during the daytime. Now you have a piece of glass. The reason you have a glass structure is because you can actually view the outside. Now you bring a blind, which defeats the very purpose of having a glass structure. In the analog world, you incur enormous amounts of costs in actually buying and putting blinds. Physically somebody has to go, particularly if it is a Burj Khalifa with 166 floors, somebody has got to go bring the blinds down if it’s a sunny day because otherwise, too much heat will come into the building and therefore air conditioning bill will go through the roof. These are all the values that get trapped in the analog product.

VG: There is a company called View Glass, which is a big company today. What they have done is put sensors and IoT in the glass. It actually senses if there is too much sunshine, and it tints the glass so that there is no sunshine that comes inside the building. So, I can actually make a presentation without any blinds and suddenly the cost of blinds goes away, unnecessary sunlight that is coming into the building is filtered out, and most importantly, the beauty of a glass building is I can actually look out and that is preserved. This company is now installing this glass with data and AI and sensors in major airports, major office buildings, etc. This is an example of taking an analog product and trying to get more value by digitizing it.

VG: And I can give any number of examples. I’m wearing a pair of glasses. This is an analog product. Now what is the value that is trapped in the analog product, you may ask? The value that’s trapped is once a year, I have to go to an ophthalmologist, who will have to give me a prescription for a new pair of lenses. Then I have to go to an optician who fits the lens. Now, if I can somehow, today it’s not possible maybe nobody has thought of it, if I can put sensors on my eyeglasses such that on a continuous real-time basis it is adjusting the glass as my eyesight deteriorates. What is the value I can unlock? The value I can unlock is I avoid all the costs of going to an ophthalmologist, all the costs of going to an optician, all the costs incurred in buying new pairs of lenses. More importantly, today in the analog world, I correct my eyesight only once a year, but my eyesight is actually deteriorating every day. Whereas if this becomes a digital product, then my eyesight automatically gets corrected as it deteriorates. This is the power of thinking data and AI into your physical product and thereby unlocking value.

Kirk: So you and Venkat are both business school professors. Tell me about how you approached writing this book. I noticed that it’s it’s very specific. It’s not like a sort of typical strategy book that talks in sort of these broad terms about initiatives and ways of thinking about strategy, but you actually have some specific tools and ways of thinking about digitization. How did you go about writing this and making it such a powerful book?

VG: I think people are tired of hearing the word digital. There are umpteen books on digital—we’ve been talking about it for the last 25 years. Why? One more book is the question. Therefore, we said we have got to give an actionable framework. I will put the existing books on digital into two camps. Either they talk about technology. So there is a book on AI which will talk about machine learning, deep learning, things like that. optimization—managers are not interested in understanding the technology itself. There is a second group of books which really focuses on digital as an efficiency tool. That means they are simply digitizing processes. There is no book out there which tries to unlock how you can digitize the product itself. I gave the example of the eyeglasses. Now you are digitizing the product itself. There we said we should go beyond saying digital can change strategy to actually give them an actionable framework by which you can do it. Therefore, what we did was we studied how the digital giants destroyed the consumer industries in the last 17 years and developed our strategy principles, which are not just simple blah blahs, but they are actually actionable.

VG: Now, how did we do this? We took a look at a company in the consumer sector, say Sony Music. I used to teach a case in strategy, Attack on Sony Music. And when I taught that case 25 years ago, we compared Sony Music with what Columbia Music does, what Virgin Music does, who has the better competitive advantage, and what Sony Music should do next. But that is based on physical assets concept because Sony Music was selling albums and selling CDs. Now the problem, even though they were B2C brands, their processes stop B2B, their processes stopped at the retailer level. So Sony Music only knew how many albums they sold, how many CDs when they went, they sold. What Spotify did was they were able to observe what music the individual is listening to in real-time. The moment Spotify was able to do that, Spotify changed the strategy in music. So that is the concept we came up with called data graph. Spotify has a music graph. Google has a search graph. Netflix has a movie graph. Because the movie studios only knew how many movies they produced, perhaps the box office collection. But a movie studio did not know what movie VG was watching in real-time, whereas Netflix knows. The moment you know product as consumed as opposed to product as used, strategy changes.

VG: Google was the first company which came up with this data graph idea. They called it the search graph. We give it the generic name data graph. Now for a moment, think about Google as nothing more than a physical library. That’s what it is. You walk into Howe Library in Hanover—that’s a physical library. It’s got books on history, books on business, books on travel, books on fiction, books on nonfiction. It’s categorized into different. Now, if I went to a physical library and suppose I took a look at ten history books and decided to withdraw one, the librarian knows the book that I withdrew that day. He does not know that I looked at ten books. Suppose I come a week later and take a look at look at 50 travel books and withdraw the book to travel to Uzbekistan. The librarian knows I withdrew the Uzbekistan travel book that day. He doesn’t know I looked at 50 books. Certainly, he doesn’t know. I came a week earlier and was looking at history books. This is what Google is able to do. Google is able to keep track of every interaction the customer has with their products. By the way, on the day they launched, Google created 50 million different data attributes. By 50 million different data attributes what I mean is jaguar as an animal will be an attribute. Jaguar as a car would be a separate attribute. Jaguar as a football team would be a third attribute. They had 50 million separate attributes. Today they have trillions. They are able to find relationships across these attributes as I do a search. Then they’re able to compare my search with 3 billion others who are searching on Google. That is the power of data graphs. And we use the word graph because graph is what we learn in high school algebra. A graph is nothing more than a relationship between two variables, x and y. And if there are only two variables, I can take a paper and pencil and draw a graph. But if I had a trillion variables, I couldn’t do it with paper and pencil. You need a really, really powerful machine learning. Deep learning. That’s what is available today.

VG: So, a data graph is at the core of how digital giants destroyed the consumer industry. I say that is the strategy industrial companies should adopt, because when you take a look at industrial companies like General Motors as an industrial company, even though it is a B2C brand because individuals drive the car, their processes, just like Sony Music stop B2B. General Motors only knows how many cars they sold. At best, they may know the total miles I drove in the car when I bring it to the dealer once a year. They don’t know how I am driving that car every day. That’s what Tesla is able to do. Tesla is an example of a company which has fused physical and digital because they fuse physical and digital, I own a Tesla.

VG: Tesla knows exactly how I am driving the car every day. I had the good fortune of meeting Elon Musk a decade ago. He said something which was very powerful. He said I’m not going to produce automobiles; I am starting a data and AI company which will also produce automobiles. The moment you think of yourself as a data and AI company, the business definition changes. General Motors thinks of itself as in the automotive business. Elon Musk thinks of himself in the data and AI business. Therefore, they have now gone into auto insurance. Now, how can they compete in auto insurance better than the auto insurance company? The incumbents like Hartford Insurance and Travelers and so on and so forth. Because he tracks how I drive. I am almost going to be 75 years old. If I am a careful driver, then he knows my risk is low. He can give me a low insurance. Whereas Travelers pools the risk based on demographic classifications. They may say a 75-year-old guy, he must be risky. So they’ll charge me high insurance. When you have a data graph, you are able to give hyper-personalized recommendations in real-time. This is what digital giants are able to do in the consumer sector. That is what industrial companies should be able to do. If they can observe product as consumed, not product as sold.

Kirk: To talk about an example from, say the purely industrial sector, like Caterpillar or John Deere that makes tractors. How does the data graph apply to objects that are working out in the environment?

VG: Let’s take John Deere. Historically they were a classic physical asset company. In the 20th century, that’s how they won. They made tractors bigger, faster, more reliable, higher quality, etc., and therefore in the 20th century, their processes stopped with BTB. They knew how many tractors they sold to the dealer. That’s all they knew. Even though the tractor is actually used by the farmer, they never really bothered collecting the data. Today there are smart industrial companies. And let’s take an example of how the value that is trapped in the analog world they were able to unlock when they became a digital industrial company. Imagine a 2000 acre farm. A farmer plants in the 2000 acres. That’s a large tract of land. When you plant, there are also weeds. Now if you are a farmer in the analog world, how could you have removed the weeds? One thing you can do is to employ hundreds and hundreds of thousands of laborers and ask them to go row by row by row, identify the weed, and pluck it. That is ridiculous because we don’t have hundreds and hundreds of thousands of laborers in the U.S. but even if you had them, it will take months to do it, it will be expensive, and it may even be error-prone. A second approach in the analog world would be blanket spraying. So I don’t have hundreds and hundreds of thousands of employee laborers, so what I do is I take a helicopter or a drone and throw herbicide across 2000 acres.

VG: Now what is the problem with that? The first problem is weeds may be only in 5% of the 2000 acres. Now you are throwing herbicide across 2000 acres and that 5% is not concentrated—it’s distributed randomly. That’s why you throw across 2000 acres. You’re overusing herbicides. Second, a more important problem is you could kill the plant when you are indiscriminately throwing herbicide. Plants may also die. Third, it is environmentally unfriendly when you throw so much herbicide on a farm. So what John Deere has done is they have come up with a new machine they recently introduced called See & Spray. What this machine does is it’s got sensors, it’s got computer vision just like Tesla, so this See & Spray moves at 15mph, quite fast. And as it is moving at 15mph using their computer vision, they take photographs of what they see. Lots and lots of photographs. And they have powerful machine-learning algorithms that immediately process the photograph to see is it a plant or is it a weed. Once it determines this is a weed, a nozzle opens and the herbicide is thrown to kill the weed without affecting the plant. Think about the value the farmer is getting. First of all, herbicide use is dramatically reduced because it’s very, very precise. Second, you don’t kill the plants inadvertently. Third, environmental sustainability is preserved. This is an example of how new value gets created from the same old equipment. If you can embed data and AI into it.

Kirk: For my final question, I asked VG how much urgency firms should bring to this moment of innovation by fusion.

VG: I think they should bring a tremendous sense of urgency because this is what happened in the consumer sector. The industrial and the consumer incumbents ignored the digital threats, saying, hey, they’re not going to affect us. They’re only changing the process, but they change the products too. And therefore most of them vanished. The same fate could happen in the industrial sector, but a few things are in their favor. Unlike the consumer sector, the product will still be an important part of fusion, whereas the albums and CDs just disappeared. That was the problem there. Whereas that’s not going to happen here. Therefore, if I’m John Deere, I have a couple of strengths because the tractor is still important. I know how to design a tractor. I know how to make a tractor. I know how to make it reliable. That’s one. Domain expertise. That’s important. Second, relationships with the customer. What happened in the consumer world? The reason why Spotify and Netflix and so on and so forth destroyed so easily the incumbents is individuals gave permission to Spotify and Netflix to collect data because we quickly checked the box I agree so that we can get to the app. So without even knowing, we gave permission to Facebook, Google, and all these companies to collect data on how we live, how we pay, and how we watch movies. That’s not going to happen in the industrial sector, because first of all, there are few in number because the the number of farmers who buy tractors is fewer than the number of individuals. And the moment you sell the tractor, the tractor belongs to the farmer, not to you if you’re John Deere. So you’ve got to go and seek permission from the farmer to collect data, and the data cannot be collected just with smartphones. You need sensors, you need IoT. You can install all of that and you got to get the farmers’ permission. And the physical product is still important. So given all of this, the change will be a little slower in the industrial sector as compared to the consumer sector. But in 17 years they destroyed so many industries. So I would say, yes, you got to move with a sense of urgency, but also move with a sense of confidence because you have some advantages that the digital players don’t have. You know, they don’t know how to make an aircraft engine, they don’t know how to make a tractor. They are important, too. Therefore, I visualize a world where there will be an ecosystem of part players between industrial companies and digital companies forming an ecosystem. The only question is who will orchestrate the ecosystem? Who will take the leadership role? Will the digital giants take the leadership role and push the industrial companies into a commodity which just produces a car? Or, industrial companies will take the lead role and bring digital companies into the ecosystem is an interesting question.

Kirk: I’d like to thank my guest, Vijay Govindarajan. You have been listening to Knowledge in Practice, a podcast from the Tuck School of Business at Dartmouth. Please like and subscribe to the show and if you enjoyed it, then please write a review as it helps people find the show. This show was recorded by me, Kirk Kardashian. It was produced and sound-designed by Tom Whalley. See you next time.