Towards Better Decision Making

Look at your typical day. There is one thing that you constantly do from the time you wake up in the morning to when you go to bed at night: take decisions. These range from seemingly inconsequential things such as opting to have cereal instead of a bagel for breakfast, to big-ticket decisions such as evaluating a possible takeover target for your company.

Phil Rosenzweig, a professor of strategy and international business at IMD, has taken a hard look at the art and science of decision making in his upcoming book Left Brain, Right Stuff. He focuses, in particular, on the most consequential decisions of all: strategic decisions, ones where you need the ability to inspire others, manage risks, transform outcomes, and also outperform rivals. These decisions require different kinds of mindsets.

In this interview, Rosenzweig explains why strategic decisions are different and how one can make better decisions:

Q. You say that decisions should be put in different kinds of categories, where are you coming from and why?

A. The word “decision” used to mean many different things. Different decisions call for a somewhat different approach, a somewhat different mindset. So much that has been written recently has drawn on really good work in cognitive psychology and has told us that we are prone to certain predictable errors or biases, and we should find ways to avoid them. That’s absolutely right for some kinds of decisions, but it begins to break down when we apply those lessons to other kinds of decisions. One of the kinds of decisions that has been studied a great deal has to do with consumer choice. If you think about a consumer decision, you can either choose to buy this box or that box, but you’re not in a position to actually transform or improve what’s inside the box. Or if we think about investments, you can buy a share of Google or a share of IBM, but you’re not in a position to improve either one after you buy it. You just buy it or you don’t buy it.

There’s something else about those decisions. When you go shopping, you’re trying to make the best decision for you, for the enjoyment of you and your family. Or if you’re investing, you want the best returns for yourself. But you’re not in competition with other people, so most decisions that we commonly think about involve no ability to shape or influence the actual outcome and no need to outperform rivals. For those kinds of decisions, the lessons we’ve been told are absolutely right. You want to be aware of certain biases and avoid them. But many decisions, including some of the most consequential decisions we make, are very different. You can transform and influence outcomes and you do need to outperform rivals. Those kinds of decisions are fundamentally different.

Watch the video below:

Q. You have a model of categorizing different decisions. Can you walk us through the thinking behind it and how you arrived upon the different categories?

A. The way that I thought makes a lot of sense for most decisions that you encounter is according to exactly two dimensions. The first dimension is “Am I making a choice among things I cannot influence or a judgment about things that I cannot shape?” Or “Am I in a position to control outcomes?” So I use “control” in that sense: “low control” versus “high control”. And that would be one dimension. The other dimension is “What I’m seeking to do to do well in absolute terms? Or am I trying to do better than somebody else in relative terms?” I would say maybe 90% of what has been studied in decision research experiments fits very nicely into this first quadrant where I cannot influence outcomes: I can make a choice but I can’t change the thing I’m choosing and performance is absolute. That applies to consumer choices, investment decisions, lots of things like that.

But there are three others if I just think about the ability to shape outcomes. In many tasks, for example, maybe I’m trying to learn a language or trying to play the piano, I don’t make a choice. Do I want to be a good piano player or not a good piano player? I actually have to achieve that. And when I have to achieve something, positive thinking matters, attitude matters. Sometimes believing I can perform better than I might be able to has a way of inspiring me or encouraging me. So the thing that we’ve been told to avoid—biases, optimism, overconfidence—makes a lot of sense but I cannot actually transform outcomes. When I can, those things can be useful.

Now if I think of the other dimension: if you and I are in competition for something, it’s not good enough for me just to do well. I need to do better than you, and you’re trying to do better than me, which really is the definition of strategic thinking. Many examples of strategic thinking involve game theory or other sorts of competitions, which are very elegant ways of modeling competitive interaction. Typically in order to emphasize that dimension of competitive interaction, they’ve dropped, they’ve set aside the dimension of shaping outcomes.

What’s really interesting is, if we come to the fourth quadrant which combines, on one hand the ability to transform outcomes, to exert control, and on the other hand, the need to outperform rivals. The decisions you make like that are oftentimes the most consequential of all. Those are strategic decisions by leaders in organizations, corporations, government, military and sports. Anytime you must produce better outcomes and do better than rivals, you’re in that fourth-quadrant. The great paradox is, those are the most consequential of all decisions, but they’re the ones that have been the least studied. Why? Because they’re not as easily amenable to neat laboratory controls and replicability. So there’s this great paradox going on with the things that are most important have been least studied.

Q. Let’s focus on the forth quadrant and the kind of decisions made there. Is there a certain kind of mindset that is needed to crack that?

A. When you crack something, you get the exact answer. I think what’s really complex about those decisions is you never quite crack them. What you need, I think, is not one mindset, but two. This is what I try to get in my book, Left Brain Right Stuff. There is a time when in order to do good strategic thinking, you want to do the most careful rational job. “Who are my rivals? What do customers want? What might my actions trigger in terms of reactions? What would I do about that?” All that calls for a mindset of very thoughtful analysis and dispassionate thinking.

Game theory is brilliant there. But you also need something else and that is when you are managing, when you are leading others to achieve things, when you’re leading others to achieve things that may be unprecedented, that they might not believe that they could achieve. You need another kind of mindset which I call ‘The Right Stuff’, or the ability to inspire and motivate people to go beyond what has been done before. What you see in many great leaders is the ability to have not just the one mindset but the two to understand that a couple of very different approaches are needed to be able to move back and forth among them.

Q. Can you give me an example of that?

A. I’ll quote former GE CEO Jack Welch, a very well-known executive, where in some of the choices they made about entries into markets, about sales of underperforming lines of business, closing some buying others, he had this sort of dispassionate rational left brain approach. Welch would tell you that, “you may have this dispassionate point of view but when you’re communicating to people in your organization, you don’t want to tell them ‘well the odds of success are this percent or this is the risk we face’, you need to imbue in them a confidence to achieve perhaps at a level that is a bit exaggerated.”

Now if we come back to the first element “to watch out for overconfidence”, well, wait a minute. When you can actually get things done to have a level of confidence that exceeds what is objectively warranted is in some ways useful, and in a competitive setting it may be more than useful that it may actually be essential. That’s why ultimately it’s not one mindset, it’s a few. Most important is the versatility and the ability to move among what we have often thought that was somewhat conflicting to your contradictory ways of thinking.

Q. The popular notion is that most people skew toward one kind of mindset, but what you say requires them to do many things, think in different ways.

A. Yes, I don’t deny that most people find one works more comfortable than the other, and that’s probably why most people are not great strategic leaders. They might be very good analysts but not very inspirational. They may be very inspirational, but not very good at their ability to think in competitive terms. Certainly what they don’t need is just to be reminded over and over: “Hey, you know people have a tendency to make certain kinds of errors.” Okay, fine, that’s good for certain kinds of decisions, but for the most important leadership decisions involving competitive circumstances and the need to inspire and motivate people to do more than has been done before, you really do need a mix above a couple of different approaches.

Q. While making decisions, there are all these biases and errors, and different kinds of issues emerging because of the way the problem is framed. So the way the question is framed might force me to think in a certain way instead of another way. As an individual, is it possible for one to be conscious of different biases and condition oneself to rise above them?

A. The answer is yes, but not perfectly. There are certain kinds of biases that seem to be so strong that even if one is aware of them and even if one can take corrective action, it’s always a bit of a struggle. I’m referring to work that others have done that shows that there are certain kinds of ways to make judgments and choices that are just so natural and so quick that it’s hard to guard against certain errors completely. The evidence does show that if one is sensitized to certain kinds of mistakes that we are likely to make and, therefore, know to not follow the immediate impulse but to frame things differently to be aware of traps. When people are trained to be aware of certain things, they become less vulnerable to those errors.

Q. You’ve been talking confidence, and how overconfidence can actually be a good thing. That goes against the popular notion. Look back at all the companies that failed. That failure is often attributed to overconfidence. How and when does too much confidence become a good thing? Is there a point at which it stops?

A. “Overconfidence” is a word that has been used, overused, misused. It’s been used to mean so many different things that I think we don’t really know what it means. The way that word is used most commonly in our everyday conversation is: it is something that is attributed when something goes wrong. A company failed, “Oh, the CEO was overconfident.” A team lost, “Gee, I guess they were overconfident.” It’s a great storytelling device.

The good news is in decision research, they didn’t fall into that trap. In decision research, overconfidence means something else: confidence that is greater than what is objectively warranted. So, if I believe that I can run 10,000 meters in 40 minutes, and I’ve never come close to that in my life, and I don’t, we might say that’s overconfidence, any level of confidence that exceeded what was objectively warranted. However, we also know that if every time I go out to run 10,000 meters, I say: “Well, gee, every time I’ve run, I’ve run it in 50 minutes, I guess I’ll run in 50 minutes again.” I probably will never run a lot faster than 50 minutes. So we also know from a lot of things including athletics, that if one has a level of confidence that slightly exceeds what is subjectively warranted, that’s not a bad thing. That’s probably how you’ll improve your performance. Now in that case, is that overconfidence or not? By one definition, it is overconfidence. By another, it’s actually an appropriate level of confidence. And that’s just in absolute terms.

Now let’s imagine, I’m racing 10,000 meters against a whole lot of other people, and the point isn’t to do better than I’ve done in the past, it is to do better than all of them. The person who wins that race has very likely had a level of confidence that day that exceeded what was objectively warranted. It actually is essential from a competitive perspective. Again, overconfidence means excessive, excessive compared to what eventually happened. That’s the way we usually use the term, but I don’t know if that’s valid, excessive compared to what I’ve done in the past, excessive to what I might do today, or excessive to what I need to beat you who’s also trying to beat me. Every one of those is a slightly different way to think about the word “confidence”.

Q. You also talked a fair bit about “control”, and exaggerated “control”. When does it become a good thing?

A. “Control” is another word that can mean a few things. I say: “Well, this is a very controlling person.” That’s not a nice thing. But I mean control differently. Control is the ability to influence outcomes, to shape outcomes. So if I’m running 10,000 meters, I’m the person who has to put one foot in front of another. I might not reach the score I want or the time I want, but I have control on that. There are other things for which we really have no control.

If you look at almost 40 years’ worth of experiments, it has been found that human beings believe they have control more than they really do. We know that from experiments like picking lottery numbers, or rolling dice, and we find out that people believe they can change the way dice will roll. So we have a great tradition that says people suffer from an illusion of control. But look at each of those examples, and you have zero controls when it comes to picking a lottery number or rolling dice. To the extent you believe you have any control, by definition it’s excessive, because in reality you have none. One of the problems is we’ve done experiments like that where you have no control. But if we change the experiment so that there’s a range of control that you have, we find something very interesting. Some people exaggerate, but some people actually underrate what they think they can control. When you have no ability to exercise control, you will exaggerate. When you have a high ability, people actually underrate the control that they can exert.

Far from a systematic bias of excessive control, what you really have is a reversion to the mean. We tend to be in the wrong on both extremes. But this is a very important finding. It’s not mine, I quoted some other people who found this. This notion that people are prone to exaggerate control is not correct. If you run experiments that are not balanced, you’ll get certain results. If you balance your experiment, you will get a very different picture about what people can and cannot do. Why this is so important is that you may roll dice occasionally, you may play lottery numbers occasionally, but really most in your life is not exercising judgments about things about which you have no control. Much of your life is “how to accomplish this task? How I drive my car? How I write this report? How I’m going to study for an exam? How I play the piano?” In every one of those you actually not only have control, but you also probably have more control than you think you do.

Q. What are the implications of your research?

A. I am very pleased with a lot of the work that’s been done recently on decision making, but I think it is time to focus on just a subset. And because of that, there’s been this curious phenomenon in the last 10 years where a number of people have been writing to an audience of strategic managers saying: “Hey managers, why aren’t you paying attention to this research about decisional biases? You don’t seem to apply this to your work.” I think if for 10 years now we’ve been saying this and people are not applying it to their work, it’s time to stop berating them for not learning the lessons and it’s time to understand that maybe the fault is ours. We’ve not understood the nature of their decisions. I’m trying to say the reason why these people don’t seem to be listening to your lessons is you have not understood what makes their decision context unique.

What I want to do with this work is saying there isn’t just one kind of decision, there are different kinds that call for different mindsets. One of its implications for a manager is to go beyond this one word “decision” and understand there are many different kinds that call for different approaches. Two, understand what you are naturally good at and what you’re less good at and try to develop some of the things that you may be less good at. And understand the need for versatility to move among different decision patterns. What makes strategic leadership so complex is the need to think about both the dispassionate rational left brain but the ability to do more than you might imagine is achievable by motivating inspiring people. And that’s the right stuff.