GenAI Playbook Step 2

My Generative AI Playbook: Step 2 (Tech Strategy – Podcast 210)


This week’s podcast is Step 2 of my Generative AI Playbook for CEOs and investors.

You can listen to this podcast here, which has the slides and graphics mentioned. Also available at iTunes and Google Podcasts.

Here is the link to the TechMoat Consulting.

Here are the slides for Step 1 of my Generative AI Playbook:

Here is Step 2 of the GenAI Playbook:

GenAI Playbook Step 2

Here is the AI Tech Stack mentioned:



Related articles:

From the Concept Library, concepts for this article are:

  • AI: Generative AI
  • AI Strategy
  • Digital Marathons: Rate of Learning, Intelligence and Adaption

From the Company Library, companies for this article are:

  • n/a

——–Transcription below

Welcome, welcome everybody. My name is Jeff Towson and this is the Tech Strategy Podcast from Techmoat Consulting. And the topic for today, step two of my generative AI playbook. Now for those of you who are subscribers, I mean, I kind of wrote a lot of theory. I basically sent you six different articles over the last several weeks taking apart generative AI as a strategy. A lot of theory, a lot of thinking about machine learning versus how humans learn, putting into some frameworks. And basically, I’m sort of wrapping that all up now into basically three steps. So if you’re a CEO, if you’re an executive, if you’re an investor, and you’re looking at a company and you’re, you know, how is Generative AI going to play out in this company? It’s basically a three-step playbook. And today, I’m going to talk about step two. I already sort of talked about step one, but I’ll review it quickly. And then the next week I’ll put out step three and that will be that for now. So it’s been kind of a lot. If you got through all six of those articles I wrote, I’m impressed because that was way down the rabbit hole in terms of strategy thinking and concepts and I was going back to like 1920 talking about human learning studies and psychology and stuff and now we’ve got machine learning and how you build that into companies and when it’s a competitive advantage and when it’s not. Anyways that’s what we’re talking about today so hopefully this will be decently useful and applicable as opposed to strategy stuff, which I really like, but I recognize it’s not always the most useful things for someone who’s running a company and, you know, swamped. Okay, so let’s talk about it. Oh, I got to do my standard disclaimer. Did it, duh, duh. Nothing in this podcast or my writing or website is investment advice. The numbers and information for me and any guests may be incorrect. Overall, the views and opinions express may no longer be relevant or accurate. Investing is risky. This is not investment legal or tax advice. Do your own research. And with that, let’s get into the content. All right, so let’s review step one of generative AI quickly. Now, I mean, the first thing to think about is we’re talking about a couple of different things here. Traditionally, when we talk about digital strategy, that’s software, right? That’s data, that’s servers, that’s everything that’s been going on since 1985, 1990. Okay. So usually when I say data, we’re talking about traditional software. I’m sorry, if I say digital strategy, usually that’s sort of synonymous with traditional software and other types of data technology. Then we get to machine learning, AI. Okay, that’s been around for a long time too. Google has been doing this from, you know, since 2000. Predictive analysis, things like that. Okay, that’s also sort of in the same bucket or at least close to it. What we’re talking about, cheap and fast prediction being used, which usually compliments software. Okay, now we’re talking about generative AI, which really is a different animal where it’s not cheap and fast prediction, although it kind of is, it’s cheap and fast generation of content. And the simplest version of that is text. And then you get to images, then you get to sounds, then you get to generative content quickly becomes generative lots of stuff. AI agents that are writing their own code to do their own software. So they’re writing the code that is the software that runs the business. They’re editing the code. They’re writing more. Are we really still in generating text at this point? No, it clearly becomes something larger, but we call it generative AI. And it’s fine. So that’s what we’re talking about. We’re talking about a generative AI playbook as opposed to more traditional artificial intelligence machine learning. Okay, now step one, which I talked about, I gave you a couple of graphics for those of you who are subscribers that I thought were very important. And then we’re basically, you got to understand the generative AI tech stack. There’s a lot of graphics for this and Drizyn Horowitz has a really good one. I’ll put it in the show notes. You’ve got to understand the idea of a generative AI knowledge flywheel. Now this is actually true for a lot of machine learning. The idea of a flywheel, the more that something is adopted or taken by users, the data feeds back, the model gets smarter and you get a bit of a flywheel effect. Baidu talks about this a lot right now. Okay, so those are sort of two big concepts. And then there’s another big concept which is rate of learning, which I’ve been talking about for years. It’s in my books. BCG has really been on this subject. Martin Reeves, the chairman there of the Henderson Institute, which is a research wing. And rate of learning is really, really important. I’ve talked about it as a digital marathon, I’ve talked about it as a digital operating basic. I’ve sort of teed up the idea of when it becomes a competitive advantage. And really what you’re talking about is rate of learning and adaptation, which is you have an organization, a business, groups of people that can rapidly learn new things and then adapt to them quickly. That’s an important competitive dimension, which is how BCG talks about it, as a very important competitive dimension as the world becomes sort of more data rich and agile. Now, I’m going to talk about that in this step too, but the distinction that’s important is a lot of people will talk about artificial intelligence, generative AI, as building intelligence capabilities, which is important. Intelligence that you have within your organization is not the same thing as having an organization that can learn and adapt quickly. Those are actually two different things and that’s when it starts to matter when you look at like when is this a competitive advantage and not. Knowledge tends to get copied. A dynamic ability to learn and adapt does not. So we’ll talk about this when we get to moats. That’s basically the idea. Okay, so those were the three big concepts I went through. I wrote a lot about them. I don’t know why I keep apologizing for that, but I think it’s super cool, but I understand it’s kind of a lot. Okay, you get to step one. What do you do first? You’re a CEO, you’re looking at a company. This is you get to step one. What do you do first? You’re a CEO. You’re looking at a company This is where most companies are today for your generative AI playbook number one You start experimenting in your products and services and you start experimenting in your basic operations And that’s what pretty much everyone’s doing You start, you know Adobe starts putting generative AI and all its services Microsoft is doing this. Google search is doing this. Everyone’s sort of plugging it in and seeing what makes sense. And then to some degree, creating entirely new projects like OpenAI and others. And that’s all important and necessary. And no, business always begins and ends with the customer. So if your product or service is not competitive or the bar has been raised for what you’re doing, yeah, you’re dead in the water. So yeah, you always start there. Now every now and then there’s a subset of thinking within that like, and this is on my little digital superpowers list. Anytime a digital tool emerges, that creates a dramatic improvement in the customer experience or user experience. It could be developer or content creator. That’s a big red flag to me. That’s not like, “Hey, we’ve got some new tech and we’re improving our product.” No, we are making all the previous existing products obsolete. You know, that’s what YouTube did to cable television. It wasn’t just an upgrade in video watching. It was a 10x improvement. So you want to keep an eye out for when generative AI can be applied and you get a 10x improvement in the product or service because that’s a major game changer. So that’s a subset there to think about. Now the other bit of step one was, okay, you put in your products and services, most people are just experimenting at this point, and then you start putting it in your basic operations. And because this is such a clear productivity tool that everyone, you know, you use chat GPT or Gemini or co-pilot or mid-journey. I mean, you become so much more productive so quickly that, you know, you start by giving it to your workforce and you basically mandate. Everybody’s got to start using all of these tools right now and yet it’s uncomfortable and there’ll be a learning curve and a lot of it will suck but you’ve got to mandate it and And your people start using all these tools and then you start sort of systematizing that across your workforce. Okay, let’s start saving all of our prompts that are effective in a database. Let’s start collaborating on how we use chat GBT to rewrite our scripts or to redo our marketing materials or to personalize our marketing communications with our customers because now we don’t, we aren’t limited by making one graphic or one email to everybody. We can generate thousands of them very cheaply and quickly. So there’s a lot of experimentation in people, what they’re doing now, and also just basically workflows. Starting to plug it into the workflows. Generally, people start with the workflows that aren’t critical. If you put it in the workflows that customers get impacted by, price changes, communications, it’s easy to mess up and make people angry. If you put it in things like payment processing, you know, at the back end of your, you know, your finance office, okay, you’ll get more productive and there’s not a lot of high risk there. You know, so that’s usually where it starts. So we would put that under digital operating basics. Do be three would be building your AI core, which is building your tech stack and DoB6 would be human workforce, team skills, and really you got to start doing a lot of training. Everybody including management, including the board, everyone start has to, you have to understand this. Now we’re all going back to school. It’s just the way it is. You don’t want to be the one person on your management team that doesn’t understand this stuff when it becomes more and more central to so much of what you’re doing. So we’re all going back to school. So it’s a lot of human training. It’s a lot of workflow sort of putting it in. And I’m sorry, DOB3 is where you build out your digital core, which your digital in a AI core now because the AI tech stack is different. Most companies are working with cloud companies to start to access these services. And they’re buying models as a service. They’re increasingly customizing them. They’ll start bringing them in-house. So that’s kind of where we are. Product innovation and experimentation. And in the basic of ops, we’re doing DOB3 and DOB6. That is what I would call step one. That’s where most companies are today. All good. Okay, let’s move on to step two. Now, nothing I have said, well, mostly nothing I have said is about building competitive strengths or competitive advantages or modes. Right now, okay, now if you create a 10x killer product, okay, that changed it. But most of what I said is like, look, everyone’s got to do all this stuff. It’s the table stakes. We used to be traditional businesses, then we became digital businesses. Now we’re becoming digital plus AI businesses. Everyone’s got to do it. Doesn’t mean it’s going to be an advantage. Okay. Step two is about where do we start to build advantages with this stuff? And, you know, for those of you who’ve read all my Moz and Marathun’s books, or at least some of them, you know, I break operating activities and structure into two levels, motes and then operating performance. You have to compete on both. The symbol for motes is always warm up and the symbol for high performance operating activities. I always use Elon Musk because he’s crazy at innovation and speed. Within operating performance, we can start talking about digital marathons. Now, the difference here is a digital marathon is we are going to identify one or two competitive dimensions where we think it will make a significant difference versus our competitors maybe product innovation maybe Being cheaper in our factories maybe Like growth rates, I mean, but I basically laid out five competitive dimensions that I think have emerged with digital tools Which SMIL that’s the acronym Sometimes one of those five, and there could be others, if you’re better than your competitor at that as an operating activity, you can start to pull away from them in your performance. And that’s why I use the analogy. It’s a marathon. If you’re running a little faster in one of these dimensions, you can slowly pull away from the pack, the marathon, and slowly move off over the horizon. And the example I always use is, you know, one of the competitive dimensions that matters in some businesses is sustained innovation, not just innovation, but we are constantly innovating and moving forward. Well, that’s Elon Musk building rocket ships. And he doesn’t have any structural advantage there. It’s not a moat, but he is so good and so fast at this that he is slowly pulled away from the pack. And he is now so far ahead of all the other runners that they can’t even see him on there. I mean his rockets, his Raptor II and he’s landing rockets. Now, he’s just so far ahead. He’s wrapped or two in love. He’s landing rockets. He’s just so far ahead. At a certain point, when you are so much better at an operating activity for so long, your competitors really can’t do what you do. I call that the winning the marathon. Now, if he messes up and screws up, they could catch him. They could advance and match his current abilities, but he keeps running. They keep advancing, but he keeps advancing more. So the digital marathon that comes to mind when we start talking about generative AI is big surprise, rate of learning. And I’ve actually, this is the L within smile, SMIL, the I stands for innovation, the L stands for rate of learning, but really what we’re talking about is rate of learning, intelligence, and adaptation, which is if we have an organization, 20,000 people, and we’re in the fashion business, And we are just really good at studying fashion trends and studying consumer behavior within fashion, which is always changing. And we are very good, we’re like super fast at studying it and then adapting very quickly. So that when we see that people in Bali are starting to buy Ugg boots, we have more Ugg boots on the shelf within days in that specific city. But when we notice that people in Manila are starting to wear tight jeans, you know, we can have our product make start to change. And not just our product means we can redesign products. We can launch new products like speed of learning and adaptation in fashion tends to be a pretty powerful dimension competitively. She and in Timu have taken advantage of this. Zara and H&M used to be the fast. They used to be called fast fashion. We used to study that business model in business school. Well, they’re much slower than she and they are much slower than a lot of these digital companies who have up the bar. Now they call them ultra-fast fashion. Okay, so we’ve talked about that before, that’s fine. But think about rate of learning when we start to add generative AI to the mix. Suddenly, we can learn if our customer who’s on the phone or on the chatbot has a problem and if they’re happy. We can respond as the chatbot in real time to constantly solve their problem and not just that, but to make them not angry. And that’s one of the outcome metrics these chatbots look at. They measure your emotional sentiment. So the chatbot is learning about what you want, what you need, and are you happy in real time? And it’s adapting in real time by giving you different types of answers and responses to slowly nudge you to the outcome it wants, which is the problem solved. Jeff’s happy. Thank you very much. That is a level of learning and adaptation we haven’t seen before. And the way BCG talks about this is like rate of learning used to take months studying new products. Then it was days, machine learning can do rate of learning in milliseconds. It can watch what you’re doing online. It can watch it look at the products you’re looking at at Amazon and it can start to adapt not just the product mix but every aspect of your user experience in milliseconds. It can start to change prices. That type of rate of learning and adaptation is something totally new, which is why machine learning and generative hours so important in this category. So that’s what we’re talking about. Now, the other part in there is intelligence. Is the machine learning getting smarter and smarter about me? The more I interact. So one, it’s learning and it’s adapting about what I’m doing looking on Amazon. Is it building an intelligence framework about me specifically? And the answer is yeah. Is it building an intelligence framework about consumers like me? Yeah. So you also have this sort of baseline of intelligence that’s being built. And the difference there, which is why it matters for moats, sometimes that intelligence framework is very static. If the machine learning learns how to translate English to French, that’s a base of knowledge of intelligence that doesn’t change that much. But if it’s learning about what Jeff likes to do today in terms of videos to watch any commerce, well, that could change day by day. You know, what I was watching on TikTok a week ago might not be what I’m watching today. So sometimes the knowledge base is very short term and sometimes it’s more permanent. You tend to be able to build the competitive moats in areas where it’s not permanent. If it’s a static knowledge base, people copy it like translation ability and grammar are basically commodities now. But rapid adaptation about what I’m watching on TikTok keeps changing. Right? So that’s how I kind of think about it. I think about rate of learning and adaptation. When is that an advantage? And then I think about the intelligence and knowledge base. When is that a real moat? And it’s usually when it changes, not when it’s static, because it becomes sort of common knowledge in a commodity. Mapping is like that. I use that example a lot. If you have the knowledge of what the traffic patterns are in Beijing today, that is very valuable knowledge. The traffic patterns from yesterday don’t help you very much in terms of the consumer experience. So that is an important knowledge base, but it degrades very, very quickly. Okay. So digital marathon, this is the first place when I look at where can I build a competitive strength or advantage with generative AI. Number one on my list is digital marathon in rate of learning adaptation and sometimes intelligence. Well, sometimes for all of them. Okay. That’s sort of the first thing to think about. The second thing to think about. I’m not going to summarize this too much, but I basically did a lot of writing. I wrote you two articles on how do humans learn as a digital marathon and how do machines learn as a digital marathon? And people are writing about human learning for almost 100 years. Because it turns out human beings, the more we do something, we get better at it. Very obvious. The more you speak Spanish, we get better at it. Very obvious. The more you speak Spanish, the better you get it Spanish. The more you work on the factory assembly line making aircrafts, as a worker, you will get more efficient. You will get smarter. Your quality will go up. And it’s not just individual humans, it’s teams of humans. You put a team of people assembling aircraft, very labor intensive. Over time, they will get cheaper and better and faster. So we sort of look at learning curves and we map them proficiency in learning, which would be the y-axis. How much better, faster, cheaper are you getting? And then the y-axis would be experience cumulative experience time spent the assembly line worker who has been assembling aircraft for three months is much faster and better than the person who’s been there a week now those curves the shape of the curves a matter because some of them keep going up but a lot of them flat line if we’re looking at a learning curve that flat lines, we are probably not going to be able to build a competitive advantage there because it doesn’t keep improving. It doesn’t keep increasing. Therefore, everyone’s going to catch us. But if it’s an exponential curve that keeps going up or more likely a linear curve or more likely an S curve with a very, very high flat line. We can stay above ahead of our competitors for a long, long time. That’s a marathon. Now that would be for humans and if you want to know more about that, just go to the Jeff Townsend dot com and look up on the concept library. Just look up rate of learning and I’ve written a ton about how humans learn and the shapes of the learning curves are very important. Okay, but what about digital marathons in learning for machines? Now the X and the Y axis, we plot these things out. That’s why they’re called learning curves. They’re curves on a graphic. The X axis and the-axis are different. And I actually spent a lot of time thinking about this and I’ve given you two graphics, I’ll put them in the show notes, one for humans, one for machines with the x and y-axis which would be the KPIs you would look for and you would check in a company. I’ve actually given them a lot of thought. So the Y-axis for machines I called knowledge and enhancement. Now a lot of this is not my language. A lot of it comes from Baidu and Ali Baba. But the KPIs we’re looking for as the machine gets smarter, we’re looking for knowledge and accuracy, We’re looking for adaptability and relevance. And we’re looking for efficiency. Now efficiency is the easiest one. Is this machine learning system chat GBT versus Gemini? Which one is cheaper, faster, lower latency? Because it turns out these things are kind of expensive. And it turns out waiting 10 seconds is annoying when you’re trying to have a conversation with the machine. So efficiency is actually a big metric that people are focused on. They’re buying all these semiconductors that are really expensive. Knowledge and accuracy is more about building the knowledge base. That’s very like, Baidu talks about this a lot, because they are building knowledge maps and knowledge bases that are industry specific. So they’re building machine learning stacks that are completely specialized for, say, factories. And the knowledge base that you would use in a factory to make it better, smarter, more adaptable, is very different than the knowledge base you would use in science, chemistry. So they’re building industry specific knowledge bases. So that’s knowledge in accuracy, efficiency, that’s another KPI. And the third is adaptability and relevance. Does the knowledge base change? Does what you knew yesterday no longer become relevant? One of the things about generative AI, which is very different than traditional software, is you can ask traditional software the same question a thousand times and it will always give you the same answer, which if it’s right is very helpful. You ask generative AI the same question a thousand times. It will give you different answers. And sometimes it can give you the right answer a hundred times, but the 101st might be wrong. If you think about humans, that’s actually how we are. We have knowledge. We can answer questions. But we in 100% accurate and we might be right today and wrong tomorrow. Math problems got it wrong today. So those three measures that they use knowledge, adaptability, and efficiency, that’s very similar to sort of my language of rate of learning knowledge, intelligence, and adaptation. Kind of the same thing. So that’s what you want to increase. That’s the y-axis. The x-axis is industrial application. How many users are using the app built on this machine learning? Okay, you’ve built your foundation model that’s specific for manufacturing. On that foundation model, people have built lots and lots of apps, developers. How many people are using them? Because that generates data which feedbacks and makes it smarter. So with on the x-axis really what we’re looking for is how widely is it used? Is it a model nobody uses or is it the model everybody uses? The more people that use it the more that they use it per day that’s gonna be the x-axis the it per day, that’s going to be the x-axis, the y-axis is then how smart is it? You also look at relevant data, how much the data is going to come from one to two places because all of these models require massive amounts of data. It could come from various proprietary sources of data that are internal. It could come from external data like sensors on the road, cameras, and then it could come from user creation. So, you know, if people are driving their cars on Baidu Maps, the mapping data is not coming from street cameras, although it is somewhat. It’s coming from the people driving their car. So does the user data, does the data you need to run the model and make it accurate? Does it come passively or does it come from user engagement? And I’ll just tee up the answer. When your data is more dependent on users using the app, that’s where you can build competitive modes, because not everyone is going to get high user engagement. Everyone might be able to buy the data from the cameras on the streets. But if you’re dependent on users to generate the data on a daily or frequent basis, that’s where your mode’s going to emerge. For the most part. And then you also have knowledge maps. Anyways, I’ll put those in there. It’s hard for me to just, and there’s different shapes of curves, and you look for the shape of the curve. Does it flatline quickly? Probably a commodity in terms of intelligence. It’s hard for me to dis- and there’s different shapes of curves and you look for the shape of the curve. Does it flatline quickly? Probably a commodity in terms of intelligence. Is it a linear curve that keeps going up? That might be a digital marathon where you can keep moving ahead of your competitors. Most of them are going to be S curves, but some of them keep going up. Anyways, okay. We’ll finish up here because I’m talking for a while. In terms of step two for the generative AI playbook, that’s really the first part of it, which is you’ve got to decide if you’re going to run a digital marathon within rate of learning, intelligence, and adaptation. And then you need to map out where within that your humans which are going to use various digital tools and or your machines can create where there’s a learning curve that is favorable to us where we can keep moving up the curve over time and stay ahead of our competitors which is basically running a marathon and we have KPIs for that and you can look at the graphics and get to and stay ahead of our competitors, which is basically running a marathon. And we have KPIs for that and you can look at the graphics and get the bit. So that’s kind of number one. The second part, which is the last part, is when can we start to build intelligence assets that are strategic and differentiated? Now for those of you who know my frameworks, how do you go from superior operating performance, which is what we’ve been talking about to having structural moats competitive advantages barriers, dangery wear, you know, you can have very you can have companies with fairly weak management and not very good operating performance But if they’ve got a big powerful moat, they do just fine. How do you move up to that next level? That’s moving from Elon Musk land up to Warren Buffetland. Okay, the linkage between those two buckets are CRAs, capabilities, resources, and assets. The example I always use is like, look, the operating activity that Walmart was very good at was opening and managing stores. That’s what their operating performance was. They did it really well. In the process of doing this for a long time, they built up a base of assets, physical assets, in this case, trucks, stores, warehouses. That base of physical assets is what eventually manifested as their competitive advantage. Their density of stores, their local geographic density, and there are some other things they did. But it’s the operating activities that generally create assets of various types. This is what they call resource-based competition. It is those resources that can manifest as moats. If you have, and they can be physical, which I just gave you an example, tangible assets, but more often than not, they’re intangible. If you have been making movies forever in your Disney, the asset you have created is intellectual property. We own Iron Man. Snow White. That intellectual property, which is an asset, an intangible asset, ends up manifesting as a competitive advantage. And if we look at our competitive advantages, intellectual property patents, big part of it. Okay. So that’s kind of the linkages. So, and as we start to do this rate of learning marathon and we’re doing this activity, you start to ask yourself what cap– and I say CRA’s capabilities, resources, and assets. What capabilities, resources, and assets are we creating over time because we’re running a marathon such that when they accumulate they may manifest as something that our competitors can’t match and The so what here is it looks like intelligence capabilities are an entirely new type of asset I Went to the Huawei Connect conference about six months ago in Shanghai and This is for their cloud business and their enterprise business. And they basically sell their enterprise services and they sell their cloud service. They have rewritten everything they do as we help businesses build intelligence capabilities. You know, they call it like intelligence everywhere. The idea that intelligence is kind of kind of be one of the key capabilities that every business must have. Now, I would define that as, you know, rate of learning plus intelligence plus adaptation. But yeah, now I don’t think it’s actually everything. I think of your Starbucks and your selling cups of coffee, you don’t really need that much. Do I really need advanced intelligence capabilities? No, the customers show up every morning, 7 a.m., they’re waiting outside for me to open the doors. But if you’re in things like fashion, as mentioned, you can see that like the ability to learn quickly and adapt could be a capability that matters. And if we are doing, you know, if we can make a standard factory dramatically more productive and dramatically smarter, okay, factory level intelligence capabilities, which is what Baidu AI cloud is selling. Very powerful. So digital marathon. Now the question is, everyone’s building these things. At what point does it become a differentiated and strategic capability that makes a difference? Because most businesses build lots of assets, factories, you know, customer bases, all of that. You have to legal department, human resources department, fine. Sometimes a couple of those become the basis for competitive advantage in a mode. It’s a very small subset. Just like digital marathons are a very small subset of a long list of operating activities that every business has to do. So I’m looking for those rare subsets where one company can’t be beat anymore and others can’t do what you do. Other rocket companies simply can’t do what Elon Musk is doing. They don’t know how to land rockets. They don’t know how to do any of this, right? You can, it’s very obvious when you see it. The trick is to spot it before everyone else sees it. Where you’re like, look, this company’s doing something. I don’t think people can match them. Well, that’s usually a small set of digital activities, marathons. And it’s a small set of capabilities, resources, and assets. No. OK, that’s kind of the point for today. We’re in step three, we will jump to here’s the specific competitive advantages you want to build number five and number eight or whatever And for those of you who are subscribers, I sent you a pretty good framework for how to think about CRAs for intelligence so I’m starting to sort of map that out, but the short version of this talk, step one, for the generative AI playbook, experiment in your products, start dialing this into your basic operating activities. DOB3, DOB6. Step two, decide if there is a digital marathon to be run here in rate of learning adaptation and intelligence that will make a difference in my business. Is this a key competitive dimension by which I can win? And for a lot of companies, the answer is no. But there is some. So look at the five and decide if this company is going to make a difference. And then you look at what the company is building there. And the reason I call it a marathon is it takes years and years. You don’t pull far ahead such that the competitors can’t touch you in a year. Now this is usually a two to three act for CEOs. I usually tell them, look, this is a 18 month to three month commitment of spending and building. So you have to understand why I’m gonna spend so much time, effort and money building this this sort of activity. I have to understand what I’m getting and the answer is you’re gonna pull away from the pack in this dimension and it’s going to make a difference with your customers and/or your competitors. You got to see the payoff. That’s part of step two. The other part of step two is start thinking of the intelligence capabilities. You’re going to be building into your organization as assets, resources, capabilities, just like you would build stores. You’re going to do a lot of that. That’s fine. But is there a short list there where we can start to see an intelligence capability being a strategic capability that makes a difference? That would be step two. Now step three, which I’ll talk about and I’ll send out an article on this will be, okay, what specific competitive advantages should you build? Where’s your barrier to entry? Where’s your moat? That’s kind of the, I always refer to that as sort of the finish line. You know, we’re going to do all these activities. Here’s all the digital stuff we’re going to do, but in two years, this is the moat you get. That’s the finish line. And once you have that in place, life gets a lot easier, you know, or at a minimum, life doesn’t get worse because the degree of competition you’re facing doesn’t keep increasing year after year. That’s the high ground, the top of the mountain, you know, the mode. I’ll talk about that in step three. But that’s the playbook. And I’ll put the graphic. I’m going to boil this all down to about five to six graphics. So step one has three graphics. Step two has three graphics. And step two will have a couple. But that’ll be it. Anyways, that was a lot of theory today. It’s take a look at the graphics in the show notes. They won’t be in the Apple or Spotify. You have to click over to the webpage. But I think the graphics, I spent a lot of time on them. I think they make it very simple. Step one, three graphics. Step two, two graphics. And then all this hand waving of me trying to describe slides to you should be more helpful. Anyways, that is it for me for today. I am… I’m having a spectacular week, like, shockingly spectacular. I’m sitting in Bali. I’m in Ubud. I’m in a villa overlooking sort of the pool and the jungle and the rice fields. Pretty amazing. And we’ve been bouncing around. We did some waterfalls in near Surabaya, sort of Eastern Java. We did volcanoes, which were really cool, Mount Bromo. And then yesterday I had one of the weirdest days I’ve ever had. Like it was so strange, where someone invited me over a friend and said, “Let’s go to this resort and we’ll have lunch and it’s a resort but they have a lot of spa stuff like they have saunas and steam rooms and they have all the stuff I’d never heard of like salt rooms where you breathe in salt air and then they had a cryo chamber which is for like people who are into ice baths well this isn’t a bath you go into a sealed room and it drops to 110 negative Celsius and you sit in the cryo room, which I didn’t do. Anyways, so we go over there yesterday and it turns out this resort is owned by Russell Simmons who, if you don’t know Russell Simmons, he’s like the godfather of rap. He was the co-founder of Def Jam Records. Like he’s the one who signed Run DMC and Snoop Dogg and I think Jay-Z. I mean if you ever watch any Netflix videos about like rap and hip-hop in the 90s, he’s all over. Like Russell’s image is huge. Anyways, it’s his resort and I didn’t know this, but I just ended up hanging out with him and chatting and I didn’t know who, really who he was. I kinda knew, oh he’s a big guy, okay, fine. Anyways, I ended up chatting with this guy for quite a while. He’s very interesting dude, very energetic. You can see why he’s so successful. And then in one of the weirder moments of my life, he said, “We’re all getting in the cryo, “everyone getting the cryo tube.” And I’m like, “We’re all getting in the cryo.” Everyone getting the cryo tube. And I’m not getting in the cryo tube. I’m not going to. I don’t have any interest in that. And he basically turned and said, “I have to say this carefully, don’t be a n-word ho.” Like I turn on these, “Don’t be a n-word ho.” And I literally stopped in my tracks. I’m like, no one has ever said that sequence of words to me before, my entire life. And he kind of laughed. He goes, “Yeah, that’s why I said it.” And he’s like, “You can’t even repeat it to people that I said it.” And I’m like, “Yeah, I can’t.” And while I was thinking about that, I’m like, can’t. And I was thinking about that. I’m like, did I just get street cred? Like in the rap hip hub community is Russell Simmons calling me and learn ho? Did I just get a tiny shred, a crumb of street cred? Because I don’t have any street cred in anything. I got no, I did nothing cool about me. I got an Australia. But I think Russell Simmons, I could put that on my resume. Russell Simmons called me a soo, so. Anyways, it made me laugh all day. Anyway, so he was real cool when I didn’t get in the cryo and he was really interesting. This he was a friend of a friend, like he didn’t care about me at all. I was just sort of a tag along, but he was very sociable and hospitable, both very interesting guy. And we went over to his place afterwards, which is next door, which is cool. And it turns out the house next to him was owned by Snoop Dogg. So there’s the resort, and then you walk across and there’s Russell Simmons house, and then there’s Snoop Dogg’s house. And there’s a there’s a path with a sign that says Snoop Dogg House. So that was my very strange Wednesday. Like, you know, I got home like that was the craziest day ever. Where did that come from? Like, I’ve worked for billionaire princes princes and I’ve hung out with famous people on jets. And I’m usually the guy nobody cares about because I’m the analyst working for the billionaire. But yeah, that was my day. Yeah, volcano. This was my week. Volcano, waterfall, got sick, and then with Russel Simmons next door to Snoop Dogg’s house. Okay, that was a very strange six days. Anyway, that was it for me. Kind of fun. I was waiting to tell somebody that story. Anyway, Ubud, which is the center of Bali, very strange place. There’s a lot of strange people here, a lot of wealthy people buying villas in the jungle and yoga centers and all this crazy stuff going on here. It’s kind of an unusual environment. Anyways, that is it for me. I hope that was helpful and/or entertaining. Have a great week and I will talk to you next week. Bye-bye.


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.


Leave a Reply