In this edition of Author Talks, McKinsey Global Publishing’s Raju Narisetti chats with Claudio Feser, a senior partner emeritus at McKinsey and lecturer of the University of St. Gallen, about his book Super Deciders: The Science and Practice of Making Decisions in Dynamic and Uncertain Times (Wiley, March 2024). Coauthored with Daniella Laureiro-Martinez, Karolin Frankenberger, and Stefano Brusoni, the book offers an innovative framework for decision making. Feser describes the cognitive, personal, situational, and emotional components that factor into decision making and outlines the role of the “decision navigator” in generating more efficient options and outcomes. An edited version of the conversation follows, and you can also watch the full video at the end of this page.
Why do we need a book about making decisions?
We wrote the book Super Deciders to explore the neuroscientific foundations of decision making. We developed a framework that should allow decision makers to make better and more effective decisions, especially complex decisions in uncertain and very dynamic environments.
Why did we decide to develop a framework, or a concept, for decision making? A few years ago, we were in the process of developing a new executive MBA program across two universities: the University of St. Gallen and ETH Zurich. We decided to include a big module on leadership skills and wondered what to include. As academics do, we conducted a review of the leadership skills that most drive leadership effectiveness, based on several available studies.
We noted that decision making was at the top of the list. Our goal was to develop the best course, so we teamed up with some of the best academics in this field to develop a new framework or concept for decision making.
Is there a science to our decision making?
We should take a step back and ask ourselves, “What decisions do we make and how do we make them?”
Humankind has been making decisions for centuries. We make decisions every day. Now, we make most decisions under uncertainty. We make personal and professional decisions based on the future, and those outcomes are very uncertain. “Who do I marry? What career do I want to pursue?” These are all decisions made under uncertainty.
Unfortunately, there is no formula or algorithm that can allow us to make perfect decisions. Work with brain scans has been very helpful in creating the evidence for us to frame decision making as a three-step process that we naturally go through.
In step one, we tend to simplify the situation. Our brains are wired to make or to represent situations in a dichotomous context. Frequently, decisions are framed by our brains as so-called “exploit–explore” decisions.
What’s an exploit–explore decision? The exploit decision asks, “Should I continue what I’m doing now and just optimize that?” The explore decision asks, “Should I change?” A typical example is, “Should I stay at my current employer, or should I change my job?” That’s a typical explore–exploit situation that, as it is presented by our brains, simplifies the problem.
In step two, the brain makes predictions about the future and about possible outcomes. For example, although it has no way of knowing, my brain might predict or assume how a job change might accelerate my career.
In step three, the brain applies seven heuristics, or rules of thumb, to make those decisions or to choose relevant exploit or explore options. Four of the heuristics are situational: the ambitiousness of my objectives, the opportunities available, the time I have available, and the social context. The latter is what other people are doing. Those will drive my decisions.
But three of the heuristics are very individual ones: my personality and inclination toward risk, my curiosity about newness, and my cognitive state. My decision making is affected by whether I feel optimistic or pessimistic.
Brain scans show that a large part of the decision-making process happens in the limbic and striatum systems. These are two parts of the brain that lie under the cortex, where the decision-making or thinking process is largely unconscious.
Generally, we unconsciously go through the process of simplifying the situation, making predictions, and applying heuristics. The limbic and striatum systems are also where our memories, our habits, and our experiences are stored. Our past experiences, biases, and inclinations often shape how we see reality, how we frame the problem, what exploit–explore options we see, and how we then work through that problem.
What exactly are super decisions?
The general framework is all about a process that is largely unconscious, whereby the brain does a lot of work using past experiences. Making “super decisions” is going through the very same process that the brain will go through anyway in a conscious, deliberate way, because it allows you to see many more options, make better predictions, and ultimately make significantly better decisions.
What is the ‘decision navigator’ you offer?
This is a six-step process.
- Identify the problem.
-
Develop multiple options on the exploit–explore continuum. So if you go back to the previous example, you can either stay at your company or quit and look for another job. Yet there are multiple options in between. You can stay at your employer but take evening courses. Build some skills that might allow you to look for other positions in the future or better your chances as you do so. Or you could commit to an 80 percent position that enables you to stay with your current employer, and maybe with that remaining 20 percent, you then try to develop an entrepreneurial venture with a friend. Or maybe you can stay at your employer but take a six-month sabbatical and try something out that you might not have otherwise.
So the idea is to develop many options that combine both exploit elements and explore elements. And that combination is really important, because introducing some exploration elements allows you to build opportunities and options for the future. That’s the second step: developing more options along this continuum.
- Identify the assumptions and predictions you’re making. For each of the options, ask yourself, “Why do I believe this is a good option?”
- Test your hypotheses. Once the assumptions are all written and clarified, they become hypotheses. Hypotheses can be tested with facts. As you look for facts, you will learn new angles, which will lead you to the “pivoting” step. Pivoting is when your options start to morph and improve as you learn new facts. After a while, your options are good enough for you to make a decision. There’s a six-step process of understanding the problem that develops options along the continuum of exploit and explore actions.
- Optimize your solutions.
- Make your decision as part of a conscious process.
Doesn’t the decision navigator slow our decision-making speed?
In fact, we make most of our decisions quickly. We rely on past experiences and intuition, occasionally. Deciding this way is actually quite efficient, doesn’t take a lot of time, and often produces “satisfactory” and “sufficient” solutions. However, sufficient solutions are not enough. For a decision about the future of a company, the future of individuals, and firing or releasing a large part of your employees, a sufficient solution is not ideal.
We know that the decision navigator produces significantly better decisions. Our research suggests that the decision navigator creates more options, significantly better predictions or assumptions, and, as a consequence, significantly better outcomes. Using the decision navigator has clear benefits.
Decisions become transparent, and you can explain them. It’s a hypothesis-oriented, fact-based way of making decisions. When implemented in organizations, it creates a situation where many decisions are not really seen as failures or mistakes. Rather, they’re viewed as learning experiences, since they are options based on tested hypotheses. Therefore, a failed initiative is not stigmatized necessarily; rather, it is viewed as “an experiment that didn’t work.” This perspective can enable an organization to create a learning mindset or a more adaptive way to look at things. Overall, people are more relaxed talking about hypotheses than they are talking about beliefs.
I see that in class: we are working with this framework and giving students complex problems to solve. And when they work in teams, using the decision navigator, the quality of debate goes up and the interpersonal relationships improve. Why? Because they discuss hypotheses, they don’t discuss “my beliefs” or “your beliefs.” And hypotheses are not very personal. They can be brought about, and you can come to better solutions.
Overall, there are three reasons why the decision navigator works. It produces much better decisions—we know that from the evidence—and much better outcomes. The decisions are explainable and transparent, and they create a much more adaptive and learning-oriented organization. That said, it takes time. We’re not advising anyone to use the navigator for every decision, but just for those decisions where sufficient and satisfactory are not enough.
You ask us not to make important decisions alone.
As I mentioned, most decisions are based on experience and intuition and need to be satisfactory, so it’s perfectly fine to take them whichever way you feel comfortable.
However, for important decisions, where sufficient and satisfactory are not enough, it is advisable to debate and discuss your decisions with other people. Why is that? When you discuss your decisions with other people, you increase the exploration radius of your options.
What does that mean? You develop more options. You see more opportunities. Your experience shapes the situation in one way, but someone else’s experience shapes the situation in another. The ability to see new angles, options, and opportunities arises when we view a situation through many lenses—not just one.
Another factor is that the brain makes predictions. Research shows that predictions become better and 30 percent more accurate when done in teams. More and better options and better predictions ultimately lead to much better decisions.
Decisions don’t always need to be made in conjunction with a team. Partners, friends, or anyone with whom you feel comfortable debating can play a role. But when the decision really matters and sufficient and satisfactory are not enough, we recommend not working alone.
You suggest decision-making teams be built using personality tests.
You want to have people think differently. People think differently when they have different personalities, but also when they have different experiences. I look at a situation through the lenses of my experience, my past, and my personality. With that in mind, here are the options I see. Now, if people all have very similar experiences and similar personalities, the exploration radius will not be very large.
If we mix up teams that have very different personalities and experiences, we will likely have a much larger exploration radius: we’ll see many more options, and we’ll see many more nuances. That is the advantage of having different mental models, upbringings, and life experiences.
That said, very diverse teams are harder to manage. People of different cultures are harder to manage than if everybody is from the same culture. It requires more empathy, more capability, interpersonal skills, and the ability to connect. [It] also needs very strong team facilitation or team leadership. Diversity really matters. It’s not simply a gender diversity or a geographical diversity; it’s really a diversity of minds that makes a difference. It also requires that everybody on the team is able to work in that diversity and appreciates diversity and different points of view, and a good dose of empathy is always necessary for that.
It's not simply a gender diversity or a geographical diversity; it's really a diversity of minds that makes a difference. It also requires that everybody on the team appreciates different points of view.
You also ask us to prepare to be wrong.
We cannot predict the future, and ultimately, some of your decisions will be wrong. But there is another element in that: it’s not only about hedging, going step-by-step, and seeing a decision more as a journey that unfolds, as opposed to a one-time event. There is another element to this.
Sometimes, you are confronted with a dilemma. Let’s say you’re facing an ultimatum: you don’t have the time and all the information that you need to make a decision, and yet you have to make a decision. Let’s say it’s a decision about letting a person go or not. Now, the question is, how do you make a decision when you’re faced with a dilemma for which there are only two outcomes—letting someone go or not—and you don’t have all the necessary information?
Eventually, you will make decisions that turn out to be wrong. It’s good that you can look at yourself after having made those decisions, that you can explain them to your friends and your parents, and that they are morally sound. That’s what it means to “prepare for being wrong.” It’s preparing yourself to make decisions by being cognizant of your values. Values play a big role, especially when you don’t have time to make a decision unfold in steps.
How will generative AI change decision making?
That’s a very hot topic, and there is no definitive answer. We don’t know what it will be in five years. But what we know now is there are good reasons to believe that AI will not “replace” but “augment” human decision making. How? Good decision making is about broadening options and working with facts.
Now, AI is super helpful in these two tasks. If you are wondering where you should vacation, you might have two or three ideas in mind. Ask AI to give you “ten tips about good places for vacation,” and give a bit of information about yourself. The output could be seven, eight, or nine options that you hadn’t thought about. So it can help you to open up your option space, and it can almost serve a little bit as a counterpart for debate, as we mentioned before, like your partner or a good friend.
The second thing AI can help you with is to conduct research and gather facts. It’s quick to point you to existing studies, analysis, or research that has been done that you might need or be able to use for your decision making or to validate related hypotheses. In that sense, it can broaden the exploration radius, and it can help you create more options. It can provide a lot of facts or point you to facts that are relevant for testing your hypotheses quickly. But ultimately, a lot of decision making comes down to very human things that are not easily substitutable.
A lot of decision making comes down to very human things that are not easily substitutable. What about your moral compass and ethical compass? What about how your personality shapes your decision making? How about your mental state?
What about your moral compass and ethical compass? What about how your personality shapes your decision making? How about your mental state? You know, today you’re inspired, you’re very positive, and you’re very optimistic. But you’ll make different decisions when you are more pessimistic. All of these are ultimately very personal, human elements.
It’s unlikely that we will see much change in this area. Also, where we see AI being applied, it’s really applied more in the research aspects. That’s part of the option development space, but not as much a part of the decision itself.
Time will tell. For now, we believe we’re in a safe and important space, because applying your moral principles, ethics, values, personality, and inclination to decision making is what really makes the difference.
Watch the full interview
