In conversation: Managing in extreme uncertainty

| Podcast

In a time of crisis, especially one as deep and unique as the COVID-19 pandemic, uncertainty is extreme and information changes daily. Most organizations realized that they lack the tools to manage effectively in such conditions. In this episode of the Inside the Strategy Room podcast, three McKinsey experts who are helping guide clients through the current crisis share the insights from their recent article, “When nothing is normal: Managing in extreme uncertainty.” Patrick Finn is a leader in McKinsey’s Healthcare Systems and Services Practice, serving management teams on topics ranging from corporate and business-unit strategy to risk management and operations. Mihir Mysore is one of McKinsey’s most experienced leaders on crisis response, helping large organizations manage operational risk, stabilize crisis situations, and build resilience. They are joined by Ophelia Usher, an expert on resilience, crisis preparedness, and crisis response and recovery. This is an edited transcript of the discussion. You can listen to the podcast episode on Apple Podcasts, Spotify, or Google Podcasts.

Subscribe to the Inside the Strategy Room podcast

Sean Brown: Patrick, you may be closest to the front lines given your work with healthcare organizations. How is this crisis different from others that you helped clients navigate?

Patrick Finn: We have long helped clients manage contained crisis events, but last year put us all in the unique position of navigating a complex set of uncertainties across the global economy. It was like being in the middle of a hurricane, with waves of all sizes and shapes across a number of issues coming from all directions, and that uncertainty remains today. In one recent week, US projections showed rising COVID-19 case volumes and the following week they were declining. Then there is the uncertainty around what the new normal will be. Crises like this tend to reshape entire economies.

The pandemic has been unique as well in that there is a large degree of daily change over a long period of time. That is very different from a typical corporate crisis where an event may have some knock-on effects but the uncertainty is largely contained to a short period. On the other extreme are crises of long duration but low daily magnitude of change, which you can see in how climate change is playing out. The COVID-19 pandemic has entailed a high degree of change over a long period. Management teams see their assumptions change weekly and even daily, and most management systems do not deal well with that level of uncertainty.

Sean Brown: Are there early indicators that can alert business leaders to start preparing for a crisis?

Patrick Finn: Some crises have early warnings, some don’t. If you think of COVID-19, some people were watching their supply chains, particularly in China, and had some warning and that early preparation helped them. In many cases, you have to be able to move to a crisis-management footing within hours or days.

Sean Brown: Mihir, what kind of challenges or pitfalls do business leaders need to be conscious of when they confront a crisis and need to rapidly respond?

Mihir Mysore: I remember a few years ago talking with the leader of a global manufacturer with deep engineering expertise. The company’s distinctiveness was founded on the quality of their technical expertise and they were facing a very large and visible product failure in one of their largest markets. They had a choice between issuing a recall or fixing the product on-site. The economic analysis suggested that under a wide range of scenarios, having a recall would be the wiser path, but the leadership rejected that outright. They said, “We are one of the premier engineering organizations on the planet. We are not going to admit defeat and issue a recall.”

To me, that was a classic example of optimism bias, the unwillingness to imagine worst-case scenarios, which tends to be a powerful force in many organizations. Another challenge is informational instability. In late January and early February, people were saying that COVID-19 is not that severe. Part of the information problem is that many crises have a technical heart, whether you are talking about oil wells in the middle of the ocean or malfunctioning products or viruses. The problem is only diagnosed over time and that creates significant uncertainty even as more information flows in.

Patrick Finn: Both the human brain and our management systems have a difficult time dealing with information instability. People in publicly traded companies are accustomed to managing quarterly earnings to the penny, yet last spring and throughout the year there was uncertainty on a number of dimensions. What is the virus prevalence rate in the community? When will case volumes change?

Mihir Mysore: What accompanies information instability is that the solutions to the crisis may have to be invented, not only implemented. You have to go all the way to the basic tenets of the problem you are trying to solve. Many companies are not set up to do that. Executives have a powerful instinct to act based on pattern recognition and their understanding of stakeholders and markets. In a crisis, your stakeholders, including people you may have known for a long time, start behaving differently. Employees might choose to leak information externally rather than voicing their concerns internally first. Executives may engage in turf wars, and boards may seek to take control. The pattern recognition that executives have may no longer be valid, which can lead them to choose the wrong answer. It can be helpful to create a decision-making body that, whenever it feels like you are landing on an answer, asks, “What assumptions lie behind this being the right answer? And what would it take for this to be exactly the wrong answer?”

Patrick Finn: Our normal style of operating tends to lock in assumptions. If you are the leader of a business unit, you commit to a plan during budget season and then spend the year doing everything possible to achieve that outcome. In extreme uncertainty, that may be exactly the wrong thing to do, and it is challenging for executives to admit that their direction of travel is wrong because the underlying information has changed. Creating the space within the management team to revisit assumptions, celebrate changing direction, and admit that a conclusion was wrong is critical when you are in extremely uncertain situations.

Creating the space within the management team to revisit assumptions, celebrate changing direction, and admit that a conclusion was wrong is critical when you are in extremely uncertain situations.

Patrick Finn

Sean Brown: When you don’t know what the right answer is, how do you prevent your organization from overreacting or underreacting?

Ophelia Usher: All the biases we just talked about can affect how you react. There is absolutely a risk of overreacting, but there is as much risk of underreacting. One of the key things is the idea of discovery before taking action. What assumptions do you believe in and are they still true? What is the worst-case scenario and what actions would you take to reduce the risk of it coming to pass? On the other hand, are there no-regrets actions across all the scenarios?

Patrick Finn: This requires accepting that in extreme uncertainty, there is no such thing as a forecast; there are only scenarios. I distinctly remember a colleague of mine back May 2020 making an impassioned plea to our epidemiology team that disagreed about the effectiveness of masks: “It doesn’t matter whether masks are 5 percent effective or 90 percent effective, they represent a low-cost option that is effective in almost any scenario.” Even if there is a low upside, the investment is also low cost.

Mihir Mysore: One way we frame these choices is to ask, is this a decision around stabilization or resolution? In a crisis, it is generally okay to overreact in terms of stabilization but not for resolution-type decisions, because you may lock yourself into something you cannot undo in the future. For example, most organizations tend to underestimate the need to connect early and often with stakeholders, but anything you do to stabilize is the right bias to have. Stabilization decisions, such as the decision around wearing masks, tend to be relatively easily reversible, but resolution decision—related to vaccines, for example, or whether to do a product recall—are much harder to reverse.

Sean Brown: How do you balance the informational instability with a need for quick action? You may end up doing nothing, which could turn out to be the worst response of all.

Mihir Mysore: The way to approach informational instability is to get grounded in the scenario-based thinking that Ophelia mentioned. It is difficult to predict a crisis, but it is generally not that hard to lay out two or three possible paths along which the crisis could evolve based both on stakeholder actions and resolution pathways—here is how the legal pathway might evolve, or the financial pathway or the operational pathway. Then you can usually see a set of no-regret actions. For the bigger decisions, we typically construct triggers. In other words, if this happens, what decision would we take then? And what are the early-warning indicators and conditions under which that decision becomes valid? Then you monitor to see if those triggers are emerging. Usually there are only five to seven big decisions that really matter and you want someone monitoring for the time to make those decisions.

Sean Brown: How can leadership teams prepare so they are ready to act when uncertainty hits them?

Mihir Mysore: One of the most undervalued tools in managing extreme uncertainty is the concept of an early-warning system. The term conjures up an image of people analyzing high-tech, fast-moving data, but in reality the vast majority of crises that go on to become existential do not come out of the blue—even COVID-19. When you think whether similar things have happened in the past, the answer is yes. When you think about whether, back in March or April we already knew the basic shape of the virus, the answer is, not completely, but its pervasiveness was known, at least in some parts of the world. Arguably, other parts of the world did not use those tool kits effectively, so there is a broad lesson around making sure you pay attention to the clues that matter and have a mechanism for filtering out the noise. That is the hardest part of an effective early-warning system.

Making sure you pay attention to the clues that matter and have a mechanism for filtering out the noise is the hardest part of an effective early-warning system.

Mihir Mysore

The second tool is an integrated nerve center, where a small group of decision makers, frequently in near-constant connection with board members, can make large decisions fast on behalf of the whole organization. It represents the state of emergency for the company. You need to be clear about what the decision rights are and how they shift in an emergency versus day-to-day business. In most organizations, there is a presumption that things will work fine when an emergency comes, which frequently does not end up being true, so doing some practice and simulations to build this muscle in the top team is critical. Finally, you need transparent operating principles: What values will see you through the crisis and how do they translate into practical decision-making criteria?

Sean Brown: In highly uncertain circumstances, people can often become pessimistic, and that can be toxic. How do you ensure that the organization, from a mental-health and leadership perspective, maintains its resilience?

Mihir Mysore: I once spoke with a board chairman about his criteria for choosing CEOs, and one of his comments stuck with me. He said, “I will never hire someone as a CEO who has not dealt with a personal or professional tragedy in their life.” I thought that was really important because we have seen many superstar executives who never had significant setbacks turn out to be ineffectual during crises. I am making a broad statement, but the rules of thumb matter in this case. You want people on your core team who have a track record of recovering from tough situations.

Patrick Finn: In addition to resilience, you need leaders who are able to absorb a lot and amplify it only when it drives the organization forward. Creating massive amounts of activity when you are dealing with extreme uncertainty leads to burnout, so the leader needs to absorb information and challenges as they come in, then choose the places to amplify and create pressure in the organization to drive to action.

Another practical reality is that the nerve center team needs to have the trust of the rest of management. One of that team’s roles is to make decisions quickly, which implies sometimes short-circuiting of the normal syndication and decision-making processes. If the person leading the team is not trusted and seen as having their interests 100 percent aligned with the interests of the organization, it creates challenges.

Sean Brown: As executives start to realize that a crisis is setting in, what are the concrete steps or phases that they should plan for?

Ophelia Usher: It is important to keep three dimensions in mind: discover, design, and execute. One of the critical roles of the integrated nerve center is to discover, to ask questions, to doggedly try to understand the assumptions that need to be pressure-tested. No assumption can become sacred. Those on the team need that mindset of discovery, because if you narrow down your scenarios too quickly, you can miss the evolution of the crisis.

The integrated nerve center has to discover, to ask questions, to doggedly try to understand the assumptions that need to be pressure-tested. No assumption can become sacred.

Ophelia Usher

As for designing, solutions often have to be designed from scratch. There will be elements of the response that you have never done before. It has to be an iterative process because if you get stuck, your integrated nerve center will start to break down and biases will creep back in.

Sean Brown: Can you elaborate on how to create this integrated nerve center and how it should operate?

Ophelia Usher: First, make sure you have the right decision makers in the room and that they have the right responsibilities. If they have to put out fires every day and are also responsible for scenario planning, the planning will likely take a back seat. The person doing that scenario planning should be forward-looking, not fighting fires.

Secondly, over time, you will want to democratize the process and return some things to the broader organization. For example, in the early days when your workforce had never worked from home, there was a lot of uncertainty, but after a month or two, that may no longer need to belong in this integrated group. As the uncertainty morphs, things can be removed from the team’s responsibilities.

Sean Brown: As you pointed out, responding effectively to a crisis requires speeding up the pace of decision making. How can leaders keep some of that accelerated pace after the crisis passes?

Patrick Finn: Many executives ask us that. The short answer is, you cannot rely on the nerve center to continue that pace because they will burn out. It is important not to get overly enthralled with the speed of decision making in the nerve center team. It is more about, how do you start to get that decision-making culture broadly into the organization?

Sean Brown: What type of person, and at what level of seniority, should lead the integrated nerve center?

Ophelia Usher: The leader needs to be trusted both above and below them in the organization. It needs to be someone whom the board and senior executives trust to take action and make decisions and who other members of the integrated nerve center trust will put a collaborative process in place.

Patrick Finn: Often a senior executive is essentially the sponsor. There may also be a crisis manager who is running the cadence of response and keeping the complex logistics moving, but that person also needs a trusted executive sponsor. This person is likely spending nearly 100 percent of their time on that crisis so usually it is not the CEO.

Sean Brown: As companies start to return to more normalcy in how they operate, what should leaders retain from the crisis mode?

Ophelia Usher: If I were to pick one thing to keep, it is the idea of monitoring, because that will tell you what is coming down the pipeline.

Patrick Finn: Signaling to the organization that you can stand down the nerve center is a watershed moment of celebration. But the combination of monitoring, thinking about scenarios, maintaining transparency within the management team about issues facing the organization, and avoiding analysis paralysis—those are more of the art.

Explore a career with us