0:00 Dan. All right, hello and welcome to this week's episode of the Dan Barrett show. I am the titular Dan Barrett. And this is the podcast where we learn all about how amazing people do the cool things that they do. And this week, folks, I have an incredible, incredible interview for you. Now, the background here is, I went to a conference, a sort of business conference, to learn I was spending, you know, five days out on the West Coast. And, you know, the whole deal was, like, we all had to dress in suits. It was very, you know, professional. And I was feeling a little out of place, you know, little uncomfortable. I don't typically wear suits. If you ever run into me, I look like your average person who works on the computer, right, black T shirt, black jeans, that kind of thing. In any case, I'm sitting there and I'm trying to make friends. I'm chatting with people, and I see out of the corner of my eye a very tall man. And sort of glance over and realize that this person that I'm looking at right here is Dr Alan Barnard, who is my guest this week. And when I tell you that I was shocked that no one was thronging this person, that he didn't have a group of fans surrounding him. I was I was shocked, because I have been a big fan of Dr Barnard's work for a very long time. And I just was like, I have to meet this person. So I got to go over, shake his hand, we struck up a conversation, and, you know, we've struck up an acquaintanceship ever since, he's been an incredible sort of person to hang out with and get to get to talk to, and I'm so excited to share that with you this week. So if you are unfamiliar, Dr Alan Barnard is a globally recognized decision scientist and one of the foremost experts in the Theory of Constraints, the process improvement methodology created by Dr Ellie Goldratt, which was just such a huge influence on me.
And Dr Barnard is one of the best people in the world making content and consulting and educating about Theory of Constraints today. So he is the CEO and co founder of gold rat research labs, which he established in 2008 alongside Doctor Ellie Goldratt. His work focuses on understanding why individuals and organizations make sub optimal decisions and developing tools to help them make better, faster choices. Doctor Barnard is a world class intellect. There is almost no one you are ever going to meet that knows more about this subject than he does, and he just happens to be an absolutely wonderful and fun person to talk to. On top of that, if you are interested in learning more about Dr Barnard's work, you can go over to Harmony apps.com that is harmony apps.com to see the software that Dr Barnard's research lab has been creating to help everybody make better decisions. So without any further ado, let's get into my conversation with Dr Alan Barnard from Harmony apps.com Dr Barnard, thank you so much for coming on the show. I really, I really appreciate it. Been very much looking forward to this. So thanks for giving us your time.
3:24 Thank you so much for the invitation. Yeah, it's we met at an event. We met at a conference, and I kind of like, picked you out of the audience because you're a tall man. And I was like, That's Dr Alan Barnard. I think I slapped the person next to me because I a big fan. So for those who perhaps aren't super familiar with your work, we were just talking before we hit record, that you've done a lot of different things. You've been involved in a lot of different areas of sort of research and business and consulting and software creation. So I'm curious when you meet people. Let's say you meet people at a conference for the first time and they don't know who you are. How do you tell people what you do? Because this seems to me a harder job for you than perhaps for most people.
4:12 It's also simple, because I've been very lucky and that I kind of discovered my major passion and purpose very early on my life. I grew up in South Africa, and I had an amazing grandfather. We grew up, you know, very poor. I knew from early on that if I wanted to be able to buy anything for myself, I would have to work. So I was an entrepreneur from about 10 years old, trying everything, delivering newspapers, you know, working in gardens, doing anything that could make me money. But my grandfather gave me this book called The Marquis, who's who, and it is basically a book. It was in the days where I'm encyclopedias were a thing, you know, thick book, and it just had bios of the most successful people in every field. And he asked me to read the book and to tell him what I'm reading, what am I noticing? And I said to him, What was fascinating to me is almost all of these successful people don't have ideal starting conditions. Many of them grew up in poverty. Many of them even suffered neglect and abuse. They might have had some learning disabilities, but despite all of this stuff, they became successful. And I said I noticed there were just two things in common, right? The one was a willingness to work hard. And I think that is generally true, regardless of which field that you want to go into, unless you are willing to work hard, you're not going to be successful, right? But the second thing related to how they made decisions, and I was fascinated about why is it that people make an often repeat bad decisions, especially avoidable bad decisions and consequential bad decisions, right? If it's not avoidable, there's nothing else that you can do about it, like if you had to make a decision around a forecast, right? You know, it's going to be wrong, that's just unavoidable. But there are decisions that are avoidable in terms of, you know, how frequently should I be placing orders?
Right? That's a okay. I can make it more or less frequent and see the outcome, or should I trust somebody going into a relationship? How do I protect myself? So that's something that I became curious on very early on, is understanding why people make an often repeat bad decisions, and how can we help people make better, faster decisions when it really, really matters, and that's an important thing. Because the first thing I learned when I started reading up about decisions is we make around 30, 40,000 decisions every day, most of it is in the subconscious mind, in under fully automatic mode, right? We have an illusion of free will, but most of the decisions, as you know, is in fully automatic mode. But there are some decisions that we slow down and we try to make them more in a cognitive mind. And those can matter a lot, and it's getting those right, and even more importantly, not getting them wrong. You know, basically what I call a hell yes decision is a decision that has a big upside of it works small downside, if it doesn't, you know, those are hell yes, the hell no. Decisions are ones that have a big, big downside. If it doesn't work, then only a small upside of it does. So I became really passionate and committed about studying that, and I ended up selecting what I was going to study, engineering, psychology, economics, and then did a PhD in innovation technology, of putting all of those things together, and that's what my research lab does. We study why people make an often repeat bad decisions. Based on the research, we've developed a whole range of award winning methods and apps to help you make better, faster decisions when it really matters. And of course, when I when I do that introduction, even if I'm introducing myself to an Uber driver, like, what do you do? I say we study why people make bad decisions, and, you know, try to help them make better. Oh, my God, you can use me as a case study. I think everybody tells that we all have, you know, a track record of making some really bad decisions, no exceptions. Yeah, right. Well,
8:24 very true, very true. And I was gonna say, like, the reaction is, like, oh, you need to talk to my wife or my friend or whatever. It's like, there's always, you know, or me, so no, it's interesting. People normally own up, you know, we, I think we're all aware that, you know, we all make bad decisions. You know, ultimately, it comes down to us. You know, we still selected that partner, we selected that job, you know, we selected that choice. So, yeah, it all comes back to us. Man, there's so much stuff I want to ask about, but let's stay here with the idea of bad decisions. Yeah, and you know, as being someone who is purely a, you know, casual reader of decision making literature. And in fact, I have like, a couple, like, sort of thick books that I'm kind of putting off on decision theory and all this stuff. So like, what I have noticed, though, is that people have sort of different models for explaining why people make bad decisions, right? There's some people who seem to think like, well, it's built into the architecture of the brain. We kind of naturally use heuristics, and heuristics are messy. And then you'll read other people who are like, no. Heuristics are actually great. Heuristics actually help us. It's the sort of when we try to reason things through, and we do that, you know, we unintentionally include fallacies and a reasoning that's what's, you know. So you have kind of, like these two camps where I kind of think, like, there's this sort of Blink, you know, blink was so the fourth sort of famous book that said, No, you got to use your intuition. It's actually much more powerful. And then you've got kind of the decision theory people who are like, No, you should use. A real, conscious Bayesian analysis of, well, you know, you get a spreadsheet, and there's two big camps, and they both kind of point at the other one as being the reason why people make bad decisions, right? So I'm curious, from your perspective, having worked to, not only, you know, get the literature, but like you said, you're working with real people. You're trying to give people solutions, things they can use. Yeah. Why do we make bad decisions? And particularly, it seems like we make the same kinds of bad decisions over and over. So why? Why is that the case? Do you think so we've, we've
10:32 sort of isolated four, four main reason categories. The first one is that we don't fully appreciate to what extent our emotions impact our decision making, and this is mostly when we are in fully automatic mode, right? So we tend to overreact and make bad choices or changes in our life or in our business due to some exaggerated frustration with the status quo or an exaggerated expectation of the future, right? Grass is green on the other side. So that's one way emotions are not helpful if they cause us to overreact, right? And there's a nice little formula that you can use, that I frequently use to explain that any decision has a structure of e plus r equal event plus response equal outcome. So if you just think about that, what's the mistakes you can make? Is you can you can respond to an event that you shouldn't have, right because it's not in your control, nothing that you can do about it. The second mistake is you can react rather than respond. So we either overreact or we under react, we procrastinate right on something, an event has happened. We should take action as a result, but then we don't. And why is that also an emotional reason? It's an exaggerated fear of using a positive that we associate with the status quo, or an exaggerated fear of the effort or risk to make the change happen. So good example of that is somebody that smokes and can't stop smoking, right? So they procrastinate. Do they know what the benefit will be if they stopped? Yes. Do they understand what the harm will be if they don't? Yes. So it's not ignorance.
Well, what is it? It's inertia, because of some exaggerated fear of using something. And many people that smoke tend to think that smoking helps them to cope with stress, right? So the unwillingness to give up a stress coping mechanism is ultimately what blocks them, or they think it will be very hard and very risky, maybe they'll, like my mentor, Dr Ellie Goldberg, you know, sadly, passed away from lung cancer, and he kept on smoking for a long time. And you know, even though he was a genius, when I asked him about that, for him, there was a connection between his genius and smoking and his creativity, he believed that it put him into a certain state of mind, you know. So it's that unwillingness. So that's an emotional reason. So that's the first category. It just our emotions cause us to over or under react. Number one. Number two is we get overwhelmed and we get stuck. So when, when we are given an impossible goal, right? And there's so many things to do, when you actually start thinking about all the things that you would have to change and implement, that can become completely overwhelming, right? So that's a that's a second category of why we make a bad decision, as we get completely overwhelmed or decision fatigued. The third category is the one that you mentioned about. We make decisions that requires intuition when we haven't developed the intuition. So that's the way to resolve the paradox is, when should I use intuition versus cognition? Right? Is use intuition when you've developed it. Don't use it when you haven't developed it. So, so a good example of that is we are naturally extremely good at reading very fine signs and body language. You know, you get a gut feel this person is not trustworthy, right? Like, go with that. You know, treat them with very high levels of this, this trust, right? Because you've developed that instinct over very long period of time. But you know, when you have no intuition about a relationship and you fall in love in the first time. Maybe slow down your thinking, you know, write down all the pros and cons and you know, and do it more deliberately. So that's the third category. As we we sometimes use our intuition when we shouldn't, or we don't when we could have right and the. Fourth category is we have real cognitive limitations, and so for each of these categories, we've basically have developed an app on approach to help us to overcome that. So we have a decision maker app to help us to identify and challenge these strong emotions.
We have a change maker app that says, if you're trying to achieve a big goal, how can you break it up into parts and do one step at a time? We have partnered with a company that uses AI to basically learn how the top people have developed intuition in a specific field, so you can learn how they make that decision as a kind of a second opinion, right? Like, what would the top expert do with the specific case that I'm facing? Because that gives me a bit of a backup. Sometimes our bad decisions are coming from a knowledge gap, right? We ignorance, but sometimes it also comes from a confidence gap or fear gap, right? This so having that second opinion can be quite useful, and probably you'll see when you use chat GPT, right? It's like, I'm facing this issue. What should I do? Here's what I'm thinking about. And it would come back and say, Yes, I think you're right. That is the best approach, or no, there's an alternative that you haven't thought about. And then the last category of us having real limitations in terms of cognitive is we don't know how to deal with uncertainty, variability, interdependencies, complexity constraints. So those, those conditions like volatility, uncertainty, etc. And that's where digital twins come in. So so in our research lab, we've basically developed an app for each of these mindful categories. Okay,
16:46 so let's stay for a minute. I was gonna ask about uncertainty. I think we've sort of gotten there earlier than maybe I anticipated, but I think that's awesome, yeah, because I've long kind of had this sense that I think a lot of people will sort of describe their ambient emotion, just societally, culturally, as being anxious, like there's a lot of people who feel anxious, yeah, and my pet theory for this, you know, beyond, I'm sure it's really going to turn out To be, we're all inhaling micro plastics or something. It's gonna be something like that. But, like, my pet theory, is that we have situations which are profoundly uncertain in ways that I don't think we're really used to, right? Like, the example I always give is I run a Google Ads agency. We had 10 clients, I think, in a row, gets suspended all on the same day for the exact same reason, which was suspicious payments, which isn't defined in Google's documentation. And when I called Google, I was like, why are all these people getting suspended? They're like, we don't know, and nobody knew, and nobody could tell me. And it was like, it was extremely anxiety producing, right? Because we're in these environments that have these incredibly interconnected, interlocking systems that are themselves incredibly complex. And so I think it feels to most people like there's such a situation under which it's not certain what's going to happen no matter what you do, that it's easy to get rooted in place and just not make a decision, right? Not deciding, even though that's a form of decision, is it feels easier, right? Yeah, so you mentioned digital twins, which seems like your sort of maybe solution is a strong word, but your approach to decision making under situations of uncertainty, and I know we had had a conversation about that previously, so let's dig into that, like, what is a digital twin, and how does it help us in those you know, volatility, uncertainty, complexity, ambiguity, kind of situations in terms of making better decisions. So
18:53 a kind of a easy way to first of all think about it, is that there's an interesting way of defining intelligence as intelligence as having the ability to build and learn from models. So if you think about how that applies to us, we have a mental model that some of it we got through nature, through our instincts, and some through nurture, through our experiences. But these mental models help us to make predictions about what will happen, how somebody would behave. So we form expectations around that, make commitments around that. When reality and prediction or expectation doesn't match up right, we have two options. We can lower our expectations, or we can change our actions. It's that ability to build a model, make a prediction, check it against reality. Now, when it comes to very complex environments, we, like I mentioned, have real cognitive limitations, and we don't understand how all the parts combined in terms of achieving an outcome. And this. Simplest way that we explain that is through a game that my mentor, Dr Lee Goldratt, developed, which is called a dice game. And it's basically asking you a simple question. It says, you have a little business with five machines, and they work in series. Raw material comes in on the one side, or maybe people come in on the one side, and the finished product comes out on the other side, and their output is determined by throw the dice. So every day, each of these machines or processes or resources can produce on average three and a half per day, but it varies between one and six. And he asked a question, if you ran this little business for 100 days, how much would come out? Now, what's very interesting, there's no mathematical algorithm that you can use to make this reliable prediction to say at time t equal 100 tell me what the output is.
First of all, because it will vary. Every time you do it, you'll get a different answer, depending on the randomness, right? But the answer is is often so much less than what you expect. We'd think it would be three and a half per day times 100 so we'd expect 350 when you actually play the game, literally with dices and chips, right? What you discover is that if a throws a one. It doesn't matter what the rest froze, only one can go through. So rather than getting an average of three and a half, you get about half that, you get about 170 out. So that's an example of that. If, if imagine this is your business, and you have to make a commitment to your customers or to your shareholders of how much is this business capable of doing both an output and in revenues? So I can multiply that 350 by, say, $100 and say, I'm going to get, you know, 350 or $35,000 that's my commitment to expectation. And then reality is, I'm only getting half of that. That's a real show of we have very much limitations in being able to put it together. A digital twin is simply automating that manual process of throwing a dice, seeing what happens, moving it from one part to the other. So it's also like building a model, but it's using Compute to help us so we put in all the parts, we create, all the dependencies, all the constraints and governance. We press the play button and you can get an outcome. Now, today, with the technology, we can create digital twins of not just a business process or a business, but a whole supply chain that is more like a network rather than a chain, but you basically model everything. And you press the play button like a computer game, and you run this thing for a year, and you see, wow, this is what it will likely produce. And it will, obviously, if you run 1000s of replications, it will show you a range of outcomes, right? It won't just be one. There will be a range, but that range turns out to be very useful for a number of reasons.
Is when I'm making commitments, what I can say is that my business will do this year somewhere between 1.2 and $1.5 million of revenue, and the resulting profits will be between 200,400 1000 so I'm immediately making commitments in ranges rather than a specific number. But the second thing is I can do is I can look at those scenarios where I got the best out and I can isolate the conditions in the model that was behind that, and I can ask, well, how can I replicate these ideal conditions every day? And I can do the same with the worst case outcome. I can say, what was the conditions that caused the worst case and how do I avoid them? So it's like a very practical example. I often get asked to go and look at businesses, and sometimes industries that I have zero intuition, and my the question that's asked in me is, can we do better now? How do you answer that? If you have no domain knowledge, you go into mining or farming or something that you've never worked in the simplest way is you ask them for their daily outputs, and you compare the best case, the best day with the average, right? He said, The best was achieved with what they already have. How do we get the average to be close to the best? What were those conditions that made the best possible? How can we replicate that? And you do the same with the worst case. What was the worst outcome, and what were the conditions that existed there, and how do we avoid that? So that's kind is the idea. The basic idea is to think in ranges. There's best case, worst case, likely. And digital twins help us to model that range together with all the interdependencies and constraints, and help us to make. Better decisions, better commitments.
25:03 So it sounds a little bit similar to not in how you do it, because I think how the sort of problems you're approaching are of such complexity and so many interlocking parts, right? It's almost like you're spinning off 1000 alternative universes and saying, okay, in 80% of these universes, we were okay, and then in 20, you know, the world exploded, or whatever. But it almost sounds like you're doing something in effect that's similar to what like they do in statistical process control. When they make a process behavior chart and they say there's some points of exceptional variation, and there's just a lot of common cause variation, which you can kind of just, you can say there's a range of three standard deviations around the mean. And then some days when you're way ahead of that, you look at that and say, Well, what was the cause of that huge spike? And you isolate that. You either make it a standard part of the process, or you try to get rid of it, depending on what you're looking at. Yeah, I
25:59 think that's a great example. So if we take something that most people would have intuition about, let's say that you have a retail store, right? So what we can do is we can put into our digital twin all the products that they sell. We can model the demand behavior when customers arrive, what they typically buy, what's the range that they buy. And now I've built the whole model. I press the play button, and it will actually generate that chart that you've just talked about by showing me what the actual inventory is every day. Yeah? And I have a band that I want to operate in, and then there's above that it will be too much, and below it will be too little, right? So that's one way that you can use the digital twins, is just show me what's likely going to happen, yeah? And of course, shortages will trigger lost sales, surpluses will trigger additional working capital and cash. And I can see but there's a second thing that can happen to say whenever I get signal. So I have this range that I want my inventory levels to operate on, where it's good enough, right? But if it goes above or below that, what do I want to do? And now I have options. And the first option is, ignore it, treat it as noise, or treat it as signal and act. And now I have options, right? How should I act? Should I place an order? If so, yes, then for how much? Right? And if I have alternatives. Which supplier should I order it from, the one that's more costly but is more rapid response, or the one that's cheaper but takes longer? And that's a second use of this, of the digital twin is I can now create rules to manage the system, to manage the inventory, and I compare which is the best, not generally, yeah, but for my specific environment. So those are the two practical uses of the digital twins.
Hey guys, hope you enjoyed part one of this episode. It's just too good to limit to one show. Join us next week to hear the rest. You
