How To Learn Our Way To Success

success during a crisis “Darwin asserted that if anyone or anything is to survive in this world, it must learn to adapt.” – James Simon

Human progress depends on our ability to learn and adapt. All the major advances in history have involved a process of discovery, much of it based on trial and error. The graphic below describes the four revolutions that have ushered in the modern era. *

The agrarian age lasted for about 10,000 years, the industrial age for 200, the information age for 50, and the conceptual age is a mere 20 years old. We can sum up recent progress in one word: acceleration. Our challenge is to cope with increasing disruption and change. As Virginia Rometty, the former CEO of IBM, put it: “In the future, the most important quality any worker can possess will be the propensity to learn”.

Learning in a crisis

At times of crisis, our ability to learn rapidly becomes the overriding factor for success, and often of our survival. However, the specific learning techniques we apply must match the situation we face. In other words, we must learn strategically. Crises fall into two basic prototypes – Episodic Crises and Emergent Crises. As described below, each demands its own unique method of learning.

Episodic crises

These are sudden, catastrophic, and short-lived events, but they leave an aftermath of destruction. Examples are natural disasters like hurricanes and earthquakes, or occurrences like the terrorist attacks on 9/11.

The main form of learning required in these situations is the After-Action Review (AAR). It is a simple but powerful methodology, responding to these questions: What was meant to happen? What actually happened? Why did it happen? How can we prevent it from happening again?

The US military has perfected this technique and applies it rigorously after every engagement or simulation. I have participated in a number of simulations at the US Army War College and have been struck by the unwavering ability of the facilitators to arrive at “ground truth,” which is often painful to acknowledge. The process requires intellectual and moral honesty, a relentless pursuit of root causes, a “no blame” mentality, and delivery of specific plans to address any problems. The learning is then disseminated across the entire military system, thus producing a shared process of improvements over time.

Some years ago, on a visit to Tokyo, I woke up in the middle of the night in my hotel room with my bed shaking. It was pronounced but unthreatening, and in a way quite pleasant. Some beds have vibration as a feature, and I wondered whether this bed had been activated by mistake. After a few minutes the shaking stopped, and, undeterred, I went back to sleep.

The next morning, I heard the dramatic news. Tokyo had been struck by a major earthquake the night before. But my hotel, apart from some broken crockery, was unscathed. So were the vast majority of buildings in Tokyo. Japan has learned systematically from a long history of earthquakes and has devised creative methods of constructing virtually “earthquake-proof” high-rise buildings that have saved thousands of lives.

It is chilling but noteworthy that every time there is a plane crash, aviation gets safer. The industry is relentless in its quest to identify the root causes of failures. Once the diagnosis is complete, the learning is rapidly shared across the industry. The statistics on airline safety testify to the success of this approach. In 1959 there were 40 fatal accidents per million departures in the US. Today, there are 0.1 per million, despite the vast increase in air traffic.

After the appalling tragedy on 9/11, an AAR was conducted by an impartial national commission. The investigation revealed a harsh reality – the failure to synthesize early warning signals that had been identified in different parts of the sprawling US intelligence system. With the benefit of a holistic picture, this catastrophe could potentially have been prevented. The result of this finding was the creation of the Director of National Intelligence, whose job it is to coordinate information picked up in the numerous intelligence agencies and thereby keep the nation safer.

Every organization can master the AAR method of learning from its successes and failures. This is, after all, how science learns. The task always is to ensure that the value of the learning is bigger than the cost of mistakes.

To bring this down to earth, we can draw inspiration from a farmer’s homespun philosophy when his cow falls into a ditch. First, get the cow out of the ditch; then find out why the cow fell into the ditch; and finally ensure the cow never falls into the ditch again.

Emergent crises

These crises involve a relentless flow of escalating harm, where both the cause and the solutions are uncertain, and where there is no clear endpoint. The current pandemic is a prime example. Emergent crises differ from episodic crises in fundamental ways and require a totally different kind of learning.

The challenge in these cases is to make rapid choices in a fast-moving environment, with incomplete information. If the growth of the crisis outpaces our ability to adapt, chaos would ensue. This demands a process of dynamic learning and continuous recalibration as events unfold.

Learning our way through the crisis

Here, based on the best evidence, are the key learning methods to apply at both the governmental and organizational levels in an emergent crisis:

Frame the challenge

The springboard for shared learning is to frame the challenge in a clear and honest way, so that everyone can understand the nature of the problem and “the reason why” for the tough choices that will follow. Vivid metaphors and examples help to simplify the issues. Statements like, “The virus respects no boundaries,” and “When you keep yourself safe, you keep me safe,” help shape the right behaviors.

Create a dynamic measurement system

To successfully navigate a fast-moving crisis, our learning methods must involve a process of continuous assessment and re-assessment. Such iterative learning loops help us interpret ongoing changes and calibrate our decisions in real-time. The goal is to adapt to new circumstances as they emerge. This kind of fast learning is fueled by a measurement system with the following attributes:

It is selective

Measurement performs two essential functions: it communicates to everyone what is important; and it tracks progress. Good measurement systems are highly selective. They concentrate on the four to five critical drivers of performance. To measure everything is a declaration of confusion – an admission that we don’t know what is important. Think of the gauges in the cockpit of an aircraft. There is limited space. The key is to select the critical measures of performance that the pilot can absorb at a glance. Cram too many gauges into the cockpit, and you will muddle the pilot and endanger the passengers.

It is simple and consistent

Good measures make sense to everyday people and become a source of learning for them. Consistency ensures that we are always comparing like with like. Constantly changing what we measure not only causes disorientation, it arouses suspicion that we may be practicing deception.

It tracks both leading and lagging indicators

Leading indicators are our early warning system. Lagging indicators look back and tell us how we have performed. A key leading indicator in the pandemic is the rate of infection – the so-called R factor. A rate below 1 means the infections are declining and we can continue to open up. If the rate climbs above 1, this tells us an outbreak is occurring and that we need to apply the brakes and increase suppression. Watching the R factor helps us rapidly implement remedial actions before being overwhelmed. The hospitalization and death rates, on the other hand, are lagging indicators. They reflect the results of our inputs.

It measures trends, not snapshots

Snapshots tell us nothing. They are not comparative. A graphic depiction of a trend can show us at a glance whether we are heading in the right or wrong direction. Trends tell a story, reveal the need for action, and underscore the urgency of proposed initiatives. They help everyone feel part of the journey.

It disaggregates the data

Aggregated data provide nothing more than an average, and averages have only one role in life – to disguise the truth. Furthermore, it is impossible to manage an average. We can only manage its component parts, and then the average will move. To deal effectively with a pandemic, disaggregation is essential. The data needs to be classified by factors such as demographics, location, co-morbidity, behavior patterns, etc. This breakdown enables us to determine causation, isolate problems, and solve them at their source.

We tend to glorify leadership and relegate management to a secondary role. Of course, good leadership is always necessary. We need a sense of direction, clear priorities, and an inspiring message. But in an escalating crisis, that is not enough. Given the complex analytical and logistical challenges such a crisis brings, one reality emerges clearly – the need for sheer management competence. This involves not only analytical rigor, the ability to cope with ambiguity, navigate trade-offs, and make tough decisions; it requires effective project coordination, relentless follow-up, and the ability to manage messy ground-level operations.

This pandemic will inescapably cause immense health and economic damage. However, we have been forced to invent new, ingenious ways of doing things, particularly working more effectively at a distance. Many of these benefits will endure and serve us well in the future.

Twitter has already advised its employees that most of them can continue to work from home indefinitely. This will save commuting time and office expenses, reduce pollution, and, based on their experience, preserve efficiency.

We have discovered more effective ways of delivering healthcare via telemedicine. We have now seen that routine medical evaluations can be efficiently conducted via Facetime and the like. Speaking for myself, this has been a boon. To enable this, I simply needed to purchase a blood pressure monitor, a pulse oximeter, and a device that can measure my heart rhythm via my smartphone. With the aid of these inexpensive tools, my physician can monitor my vital signs without my needing to travel to a medical office and sit in a crowded room waiting to be seen. My guess is that I save about half a day by having these examinations done online. Telemedicine methods will no doubt improve even further over time and hold the promise of transforming the delivery of medical services.

Those of us who teach have had to adapt to the virtual method during this prolonged “stay at home” period. This has compelled us to reexamine our entrenched assumptions about how students learn. We have found new ways to enhance the social connections and interactivity that enrich online learning.  Of course, online methods are not the same as in-person education. But we have learned that they offer benefits such as flexibility, variety, and reach. The choice between the two models is not binary. I believe these methods will prove to be mutually beneficial. Education faces an exciting future in which we discover how to harness a dual-channel approach to create a more creative and stimulating educational experience than traditional methods have been able to deliver.

* Section headings titles courtesy A Whole New Mind by Daniel Pink


Willie Pietersen is the Professor of the Practice of Management at Columbia Business School. He specializes in strategy and the leadership of change, and his methods and ideas, especially Strategic Learning, are widely applied within Columbia’s executive education programs, and also in numerous corporations. He has served as a teacher and advisor to many global companies, including Aviva, Bausch & Lomb, Boeing, Chubb Corp., Deloitte, DePuy, Ericsson, ExxonMobil, Henry Schein, Inc., Novartis, SAP and also the Girl Scouts of the USA.