51 / Cognitive Bias and Software Development
About Wolf Alexanyan
Wolf Alexanyan is the Head of Product Management at The Software Development Company, working on lawful cyber intelligence systems. For the first half of his 12-year career, he worked as a technical specialist, after which he shifted to management. Wolf has led the research and design of such products as the eSports gaming ecosystem, air operator’s registry system for the General Department of Civil Aviation of Armenia, the world’s first levitating camera.
His main passion is studying different cognitive science disciplines to understand patterns of human behavior and thinking. He has recently finished 2.5 years of research regarding cognitive biases and published his work in two pieces. The Science of User Experience explains the importance of using our brain’s errors and biases in developing solutions for software products and development teams, and “UX CORE” consists of 105 hands-on examples of cognitive biases used in software development and team management.
He sincerely believes that humanity is on the verge of a cognitive revolution. Although cognitive biases are most used in the development of political and digital products, regular people can reap enormous benefits if they take the time to study their own biases and beliefs.
Visit Wolf’s website with his notes about project and product management: https://keepsimple.io/
Thinking, Fast and Slow, by Daniel Kahneman.
The Black Swan: The Impact of the Highly Improbable, by Nassim Nicholas Taleb.
In This Episode
Without mental shortcuts to help, there’s no way product managers could process the daily waves of information coming at us. We apply these shortcuts, called cognitive biases, to drive efficiency in how we perceive and respond to the world around us. But when we’re unaware of, or not sensitive to, these biases (that exist naturally within us, by the way), well, that’s when we make mistakes. Mistakes that manifest in our lives as product people as well as is our everyday lives as humans bumping along in our own existence.
In this episode of the ITX Product Momentum Podcast, Sean and Paul welcome Wolf Alexanyan, Head of Product Management at The Software Development Company. Fresh off 2+ years of research regarding cognitive biases, Wolf recently published two significant works in our space: The Science of User Experience, which explains the importance of using our brain’s errors and biases to develop software product solutions, and UX CORE, a compilation of 105 hands-on examples of cognitive biases used in software development and team management.
“When I was working on UX Core,” Wolf says, “I wanted to show people not just how to use the biases to relate to others and protect yourself from being manipulated, but to show how powerful our brain is.
The moment we understand how to get in touch with our own cognitive biases is the moment we seize the power to make positive changes in our own lives as both human beings and product people. As Wolf explains, the power lies within each of us.
“This is the most important thing: if we focus on ourselves and spend some time just to understand the errors that we have – instead of trying to understand the capabilities of the world and opportunities that arise – we will benefit from that much more, much more.
Tune in to hear Wolf describe the role ego plays in cognitive bias, with specific reference to the Dunning-Kruger Effect, and hear him explain why the “Blind Spot Bias” is the one that new product managers need to learn and understand before all others.
Sean [00:00:19] Hello and welcome to the Product Momentum Podcast. This is a podcast intended to entertain, educate, celebrate, and give a little back to the product leadership community.
Sean [00:01:49] Hey, Paul, how are you doing today?
Paul [00:01:50] I’m great. Sean, how are you?
Sean [00:01:52] Good. How’s your confidence level in your knowledge of the cognitive biases?
Paul [00:01:55] I’m riding high on Stupid Mountain.
Sean [00:01:58] All right. Well, let’s talk about the Dunning-Kruger effect and some other cognitive biases because this guy has built a very cool tool for product leaders to use.
Paul [00:02:06] Yeah, Wolf is definitely exposed me to a lot of the blind spots that I had in what we do to impact people and specifically the shortcuts and mental models that we build into our products.
Sean [00:02:19] Yeah, well, let’s get after it.
Paul [00:02:21] Let’s get after it.
Paul [00:02:24] Well, hello and welcome to the pod. We are excited today to be joined by Wolf Alexanyan. He’s the head of product management at The Software Development Company, working on lawful cyber intelligence systems. The first half of his 12-year career was as a technical specialist, working on network servers, dev-ops, and the like, after which he shifted to management. His main passion, to which he devoted all his free time since 2010, was studying different cognitive science disciplines to understand patterns of human behavior and thinking. You’ve most likely heard of him because of a viral tool he created called UX Core, consisting of one hundred and five hands-on examples of cognitive biases used in software development and team management. He sincerely believes that humanity is on the verge of a cognitive revolution. Although cognitive biases are most used in the development of political and digital products, anyone can reap the enormous benefits if they take the time to study their own biases and beliefs. Wolf’s website and his notes about project and product management can be found at keepsimple.io. Wolf, thanks for joining us today. Welcome to the podcast.
Wolf [00:03:25] Thank you very much. Thanks for having me.
Paul [00:03:27] Absolutely. So I want to jump right in and just hear your thoughts on the state of product management and I appreciate your take on it as agnostic of the business and artistic UX kind of aspects of the industry. I think you’ve taken a psychological approach towards it. Why is it so important for product managers to understand basic psychology?
Wolf [00:03:51] So the reason I emphasize the importance of psychology here is that behind every decision we make, it is a set of assumptions, motives, and beliefs that are, in a nutshell, a result of this or another cognitive bias that we prompted. It’s like, when you ask me a particular question, you know, the sense of the situation and the feeling that you know, it is the right moment to ask this question, all that gut feeling of yours is covered by various aspects of psychology and cognitive sciences. Like put more simply, your product should feed your audience scientifically, in this context, psychologically, and then maybe it will generate a return on investment. That’s a separate topic.
Paul [00:04:31] And there’s a challenge that we’re facing as product managers because we tend to get wrapped up in the tools of the trade. We tend to get wrapped up in the frameworks, the processes. These things are important, but what do you think are some of the things that are missing in the emphases that we’re placing on training good quality product managers?
Wolf [00:04:49] So as I see it, the problem in modern product managers, even the ones that are very famous, that they’re focusing more on the tools, practices, and the processes, you know, rather than on the essential things that should be understood first. Modern product managers can endlessly talk about focus groups, user interviews, requirement generation. But ironically, everything is much more simple. It’s like all of the topics modern managers are talking about is basically about understanding the people. And right after emphasizing that in their speeches or talks, you know, they forget about one of the most vital things, regardless of whether it’s an entry-level product manager or a senior product manager, no difference in that. When they design an AB test or a survey, they design it in such a way to avoid participant biases. They design AB tests keeping in mind the huge ton of various cognitive biases just in order to increase the quality of gathering data. In most cases, they don’t even realize that they use biases. They call their knowledge, you know, experience, but it doesn’t change the essence.
Wolf [00:05:58] Of course, I’m not saying that the existing educational materials are bad. I mean, obviously, you should talk about tools, you should talk about practices because it is the part of existing reality. I just say that it’s not as good as it could be. In order to be much more interesting and exciting, the authors of the books and podcasts should do their homework and think twice about how they make their decisions. I mean, in many ways, the answers will be so simple and obvious that they will not like to tell it in public. And I understand that. It’s better to avoid the risk and keep all management domain as a place where you need to, you know, make heavy decisions, think about your users, do product that matters, and a lot of these kinds of generic fancy phrases because it’s very safe to tell that kind of thing. I understand why it is how it is. All of us are using cognitive biases regularly. We use understanding of those biases by, you know, like our gut feeling in our emails, in dialogs with colleagues, in conference calls, in books. But because of lack of scientific information and most recent evidence as related to human behavioral research, all of us prefer not to touch this kind of topics, you know.
Paul [00:07:10] Yeah. So let me jump in there and summarize because you’ve covered a lot of ground. The thing about cognitive biases is not that we use them or don’t use them, right. They’re always present whether or not we acknowledge them. And I think what I hear you saying is product managers should understand some of the ways that the brain works. And if we could just rewind, cognitive biases, for those who are coming into this podcast thinking we’re going to talk about Scrum, this is a pretty deep topic to just jump into, so what we could maybe use as shorthand is, you know, cognitive biases, as I understand them, is it’s a shortcut for your brain. It’s a way for your brain to process information more efficiently, to take in the infinite amount of information coming into us, and use a mental model to understand the world around us. Sometimes it’s helpful and sometimes it’s not.
Wolf [00:08:02] Correct.
Sean [00:08:03] Daniel Kahneman has a beautiful way of describing it. He’s got his book Thinking Fast and Slow.
Wolf [00:08:07] Oh yes.
Sean [00:08:09] You know, and it’s the difference between the fast brain, which often makes bad economic decisions because it’s the way in which it works, versus the slow brain, which if you remember when you were a kid, your mom telling you, “slow down and use your words.” That’s the act of engaging the slow brain, the thinking brain, versus the limbic system.
Wolf [00:08:27] Correct.
Sean [00:08:28] …Where a lot of these cognitive biases actually exist. When we’re not cognizant of the biases that are naturally there, that’s when we make these mistakes. The work that you’ve done, I think it’s a great tool for the leadership product management industry just to be aware of what’s out there so that we can make better decisions and build better products based upon these known biases that we all already have.
Wolf [00:08:50] Yeah. The only thing I would add to what you said is that it’s not necessarily to talk about biases in the context of human brain mistakes, but the fact is that the reason of biases’ existence is just to, like, as Paul said, make a shortcut in decision making. So generally speaking, it is a part of our system and human beings are designed in such a way that they don’t have another choice. You know, it’s like we as a system should understand ourselves first. We build products for Homo sapiens species. Homo sapiens has an amazing system with a huge ton of subsystems working in it. As with any system, Homo sapiens’ system also has bugs, bottlenecks, sections that could be optimized, et cetera.
Wolf [00:09:37] I can bring up an analogy with the software. Let’s say you’re a software engineer and your boss gave you a task to maintain the code of a third-party partner of yours. The first thing that you would do when you see the code that was developed by another person is you will try to understand the errors in that. You will try to understand the bottlenecks. You will try to understand the architectural decisions. And based on that, you will try to see which bug you should fix, which bug is very, like, dangerous to try to maintain, and which bug should be left as it is because it doesn’t have a huge negative impact on the system. The same is about people. Like, we have a body, which is our hardware, and we have a prefrontal cortex which gives us the ability to be an intelligent species on this earth, and that prefrontal cortex consists of our software system.
Wolf [00:10:29] So it’s like, when you approach a human being as a system, you just need to read about existing biases, truly understand them, then see which one helps you in your regular routine and which one does not. Then you just need to keep the first and remove or optimize the second part, you know. The moment when people will understand their power and their ability of changing their own code will start their own cognitive revolution without waiting for third-party companies to come up with, you know, mega-fancy devices, chips, et cetera. That’s it.
Paul [00:11:02] Yeah, Neuralink was recently in the news. There’s a really interesting theme that you’ve got going on with the brain-machine interfaces, and I think that we could spend the rest of the time talking about that. But just to kind of give some shorthand to the audience if this is new territory, could you define your term cognitive revolution? What does that mean when you use that term?
Wolf [00:11:23] When I talk about cognitive revolution, I mean the upcoming brain-machine interfaces’ development and the moment when they will become something available for a majority of people. Brain-machine interfaces are the devices that are being developed now by different companies, as you said, Neuralink, there are other companies like [inaudible] that are working in the same field. The idea of those devices is to read neural activity of the human brain in initial versions. And yes, in a recent Neuralink presentation, you could see that they’ve shown that they are already able to read neural activity based on their experiments with peaks, you know, and eventually, at some point, those interfaces should not just read the activity, but understand our thought patterns, maybe even understand which particular thought or emotion of ours, our reaction, will bring to depression state or, you know, maybe they will give us advice on how to cope with our thoughts. So at some point, of course, it is very far from now, but I hope we will see that in the next 30, 40 years, the system could be able to understand the reason of regular stress by analyzing the brain activity, catching the patterns, and matching it with various busses.
Sean [00:12:36] Yeah, and that will be something all product leaders have to take into account, right. This is going to change the face of how we build all kinds of software products.
Wolf [00:12:43] Absolutely, yeah.
Sean [00:12:44] That’s interesting. You know, I liked what you said about, you know, these biases are already there, and it reminded me of the work by Nobel Prize-winning author and researcher Richard Thaler and Cass Sunstein and the concept of Nudge and how governments, you know, for years have been using these theories to inform how they communicate and how they set up their systems to impact human behavior. And we need to take these things into account when we’re building products, too. But I also want to bring it back to the goals. All of these things are really tactical. They’re all, how do you apply these behavioral economics, these cognitive biases, these concepts to the tactics that you would deploy to get a change in behavior? Our strategy is always about, how are we going to inspire people to a better future together? Right?
Wolf [00:13:30] Oh, yeah, I see.
Sean [00:13:32] Because it could be used for good or could be used for bad and we have to make sure that we’re taking these tools in context and using them to drive positive behaviors, or maybe not drive them, but inspire them, to change the language there.
Wolf [00:13:44] You know, I got a lot, a lot of feedback when I released the UX Core.
Sean [00:13:49] For sure.
Wolf [00:13:49] And there were questions from different specialists working in different domains, not only IT but like, “is it fine to use cognitive biases from an ethical and moral perspective?” A lot of people told me that it’s inappropriate to use the examples because they contain manipulation of people. I even had a short chat with a few people who had the same idea because I wanted to understand why they consider the conscious use of bias manipulation whereas when you use the biases, like by accident, without knowing the science, it is not manipulation. I wanted to see the difference in how they interpret that. And as I understood, they were not so much into the science topic, they just didn’t like the fact of seeing the pure, like, example of how they could be manipulated on some particular website or mobile application, you know.
Sean [00:14:49] Yeah. I liked that you used the example of Las Vegas in the US, right. The loss aversion heuristic is the reason why that town exists. It’s our reality, so we need to be aware of it. All the work that’s being done in the world to expose these biases and make them more available to everyone I think is going to result in a more positive world in the long run. In the short run, though, it is being taken advantage of, and that’s some scary stuff.
Wolf [00:15:12] Yeah. You know, if we go slightly off from the product management topic and we talk about changing the society, I have some other views…
Sean [00:15:20] I’m sure you do.
Wolf [00:15:21] …To that topic, yeah. You know, all that’s now happening is normal and pretty understandable. Each one of us does good and bad things. OK, we were going slightly off from the product management topic, but we will get back to that, you know. Generally speaking, my desire behind the UX Core that I released is to find some public influencers who could understand the point I’m talking about and help to make the topic of cognitive biases are much more trendy, you know.
Sean [00:15:50] Yeah.
Wolf [00:15:50] I’m telling you this because the thing is that our consciousness and skills of self-awareness could help us to make that analysis that I have been talking about in the topic of cognitive revolution right now. It’s like, we have the following choice: we can wait for 30, 40 years for the companies, like Neuralink going to implement specific chips that we will put in our head and will be able to analyze our thoughts, or we can use our multibillion-dollar device in our head and make that analysis right now. I mean, when you just sit on your own and think about how difficult [of a] day you had, the processes, that retrospective analysis that you are doing to understand what went good and what went bad during that day, it is just a piece of art. You know, you are doing something that there are not any devices in the world that, at the moment, could do that.
Wolf [00:16:45] And when I was working on UX Core, I wanted to show people not just how to use the biases to manipulate people or how to use the understanding of biases to protect yourself from manipulation, but to show how powerful our brain is.
Sean [00:17:03] Yeah.
Wolf [00:17:04] When you understand a particular bias, you can, like, reflect on your own thoughts for ten minutes, make some specific changes in yourself, change some true to false and some false to true flecks in your code, and based on that, change the quality of all the decisions of all your future life. I decided to create examples in product management because I’m a product manager. I didn’t want to create one hundred examples of cognitive biases used in regular life because I don’t want to be a life coach, you know. But the thing is that when you understand that knowledge, you can go far away from the product management, apply a huge ton of changes in your own life, and based on the errors you found, positively increase the quality of all decisions that you will make all your, I don’t know, 40, 50, 60 years of your future life.
Sean [00:18:00] Yeah.
Wolf [00:18:00] This is the most important thing. It’s like, it’s very good that we have schools and universities that give you some information, but if we focus on ourselves and spend a couple of months just to understand the errors that we have, instead of understanding the capabilities of the world and opportunities that arise day-to-day, we will benefit from that much more, much more. And that was the thing I wanted to focus on, you know.
Sean [00:18:27] So one of the things you did in the tool that’s related to this is you’ve flagged each of the biases with either product relevance or management relevance.
Wolf [00:18:37] Right.
Sean [00:18:38] Which I think was really cool. Because like you said, this is applicable in all different phases of our lives and when we’re aware of them, we make better decisions, or at least we have more control over our decisions so we feel like we’re making better decisions. But the point I’m trying to make is it applies to our leadership of product teams and the building of the products as well as how we use these biases inside of the products, and our leadership, like how are we making sure that we’re building the best possible product? I don’t think you can build the best possible product without understanding the cognitive biases that go into the teams that are building it, the leadership of the teams that are building it, and the users that are actually using the products.
Wolf [00:19:15] Absolutely.
Paul [00:19:16] Yeah. One question that I wanted to ask you is, the approach that you’ve taken is, I would say controversial in the sense that you’ve made some bold claims, like, “you can learn everything you need to know about the processes of product management in about three months.” “You can get to the level of technical proficiency in Scrum in a relatively short amount of time,” but the biggest barrier to building good products is finding those high-quality product managers. And I think the essence of what I see in UX Core and the passion that you bring to this topic is that the biggest barrier that we have is our own egos. We think we are smarter than the biases that we’re understanding. We think that by observing these biases, we can compartmentalize, put them in a box, read about them, take them down from a shelf, and make them more approachable. But what we really need to understand is that we’re just as susceptible to these biases as product managers. And learning Agile frameworks and full stack development and principles of UX design is relatively easy compared to overcoming our own egos and overcoming the self-awareness of these biases. Is that a fair assessment?
Wolf [00:20:28] Oh, yeah. You know, I can talk about this topic endlessly because the human ego is a very funny thing. At some point, the human species decided that they rule the world when they are the smartest ones. Then they divided into different groups. Then it turned out that if you have a very fancy title in a very serious IT company, you are a serious guy. If you take a look at UX Core from the perspective of working with your own ego, the funny thing is that if you really focus on the biases I describe there, you cannot tell after that that you are supreme, you know. Because look, man, you have about a minimum of a hundred biases in yourself and you cannot overcome that. You think that you have a lot of control of your life, but at your best, you will be able to control about 50 percent of your life. Most of the decisions you make are built on the paradoxes of the biases that you are prone to.
Paul [00:21:26] It’s the Dunning-Kruger effect.
Wolf [00:21:29] Oh yeah, the Dunning-Kruger effect is another thing. You know, I have seen a lot of H.R. specialists prone to the Dunning-Kruger effect, but I don’t want to go into that topic too much. But overall, yeah, the idea of cognitive biases and their relation to ego is very important. Like you can, regardless of which department you work in an I.T. company, you can use the biases to increase the quality of your own department. Let’s say you are not a product manager. Let’s say you are an H.R. specialist. You can take six or seven cognitive biases that could be an easy check of, like fragility of ego of the opponent, and based on an understanding of those biases, you can create a short survey based on which you will understand whether this candidate is a cultural fit to your company or not. I mean, the ways you could use UX Core, in theory, are endless, because it shows some basic things that all of us will be prone to, like, for the next tens or hundreds of years, you know. So yeah, the ego is a very, very big problem, a very big problem.
Sean [00:22:38] They’re interrelated. I think this bias around what Dunning-Kruger discovered, our confidence is, especially early in the mastery of any sort of content, that’s the ego at play. And then we realize, what do they call it, Stupid Mountain? Right. And then you go to the Trough of Despair and then you do gain this confidence over time. And then there’s another term for it, it’s called imposter syndrome, you know, when you actually have more knowledge then you don’t have, and there’s the balance there. But anyway, with all of these biases, I think ego is a big factor, because once you recognize it if you don’t do something about it, that’s basically ego at play. Interesting. We ask all of our guests: we’re compiling definitions of the word innovation from our space. So how do you define the word innovation?
Wolf [00:23:21] Oh, it depends from which perspective you’re asking the question. It’s like, within the company an innovation is something that is being done by going out of the existing processes of the company. It is an innovation for the company. If we talk about innovation globally, that should be, like, considered as innovation to the majority, unfortunately, innovation is something that the media says, “that is innovation,” you know. You can do a lot of innovation in your own company, but to tell the world about something that you consider as a worldwide innovation… There is a thing called Overton Window, which is basically a technique of working with society, and in order to create an innovation, you should convince the majority what is the innovation.
Sean [00:24:14] Interesting.
Paul [00:24:14] That’s great.
Wolf [00:24:14] And maybe I’m not telling very fancy things, but in a nutshell, what society says innovation is is innovation, even if you are a specialist who knows that “this is not an innovation; I saw that 10 years ago.” No. If society says that this is an innovation, that is an innovation and that’s it.
Paul [00:24:31] I think that is an answer that is just in line with the kind of thinking and observational expertise that you’ve brought to the table. I appreciate that answer very much. The last thing that I’m curious to hear your thoughts on are, what book do you think every product manager should have on their shelf?
Wolf [00:24:29] One of the books surely is a book written by Nicholas Nassim Taleb. It’s called Black Swan.
Sean [00:24:54] Great book.
Wolf [00:24:55] Yeah, it’s a very important one because people don’t understand the importance of luck in the business they run.
Sean [00:25:02] No doubt. If I could add to that, I think Antifragile is equally important, from the same author.
Wolf [00:25:08] Yes, could be. I focus more on the Black Swan because it was life-changing for me. That’s why I mentioned it first.
Sean [00:25:15] It’s a great book.
Wolf [00:25:15] Another book that you already mentioned is written by Nobel Prize winner Daniel Kahneman called…Think Fast?
Sean [00:25:22] Thinking Fast and Slow.
Wolf [00:25:24] Yeah, yeah. I read it in Russian, so I don’t remember the name in English.
Sean [00:25:31] That’s great.
Paul [00:25:31] Excellent.
Wolf [00:25:31] Yeah, those books are epic ones, you know.
Sean [00:25:33] Yeah, yeah.
Paul [00:25:33] Absolutely.
Sean [00:25:34] I have one last question for you. In the tool, you’ve numbered all of the biases. I think that’s important for reference, but they don’t seem to imply any sort of order. But if you were to introduce a new product leader to this concept of cognitive biases, what do you think would be the top three to learn, just to introduce you to the concept? Because there’s a lot.
Wolf [00:25:55] I would recommend just one because if a person will understand that one, he will read the rest, you know. It’s bias blind spot for sure. It is the bias based on which every one of us thinks that he’s not prone to biases. That’s it. If you understand that you’re prone to that particular bias, it could be the start in your discovery of the rest of the biases, and I assume that it won’t be good to separate them by importance.
Sean [00:26:23] Because the context is what’s really important, right?
Wolf [00:26:26] Yeah.
Sean [00:26:26] Because whatever the context is would define which biases are being engaged or not being engaged.
Wolf [00:26:31] Absolutely, absolutely.
Sean [00:26:33] Well, fantastic.
Paul [00:26:34] We really appreciate getting your insight on this. I think you’ve really done a tremendous service to the community in bringing these to the forefront and making them a part of our conversation today. Thanks so much for joining us.
Wolf [00:26:46] Thank you very much.
Paul [00:27:00] Well, that’s it for today. In line with our goals of transparency in listening, we really want to hear from you. Sean and I are committed to reading every piece of feedback that we get. So please leave a comment or a rating wherever you’re listening to this podcast. Not only does it help us continue to improve, but it also helps the show climb up the rankings so that we can help other listeners move, touch, and inspire the world, just like you’re doing. Thanks, everyone. We’ll see you next episode.