Success Is Good. Failure Is Bad. Any Questions?

Failure is a dilemma. Even if your corporate culture is intolerant of failure, leaders know that the result will not be a failure-free company. It will be a culture that hides failures and cannot learn from them.

On the other hand, if your company embraces “Fail fast, fail often,” it will not pause to investigate basic mistakes that could easily have been corrected.

Failure as a Binary Concept

Amy Edmondson is the Novartis Professor of Leadership and Management at the Harvard Business School. She has a useful perspective on failure (2023).

Binary thinking places concepts into either/or categories. Failure is bad. Success is good. For example, an instinctive aversion to failure leads to wasting board directors’ time with “dog and pony shows” focusing on success. Meantime, the board remains ignorant of festering problems.

At the supervisory level, viewing failure as a binary concept encourages micromanagement. And micromanagement leads to high levels of resignation among young workers. (Banerjee & Bannerjee, 2023).

Level 1 Failure

Edmondson suggests reframing failure and your management of failure. She describes three different failure categories we will call Levels 1, 2, and 3. And we will describe a fourth category.

Level 1 failure is a “basic failure.” Basic failures involve errors in well-trod terrain. Time, energy, and resources have been wasted. A common example might be forgetting to lower the garage door this morning when driving off to work.

Managing level 1 failure involves forgiving the person who made the error. Level 1 errors are part of being human. Focus on what can be done to reduce the probability of the same error being committed in the future.

Notice we said, “reduce the probability” rather than “make sure it can never happen again.” The idea is to encourage realistic failure reduction and not idealistic failure elimination.

Help yourself and your team articulate issues like attention to detail, assumptions, and overconfidence. You want to create psychological safety for team members to discuss Level 1 errors. Board directors might thank CEOs who provide them with information about Level 1 errors plus plans to reduce the probability of the same error occurring again. This type of information is more valuable than dog and pony shows.

Level 2 Failure

A Level 2 failure is a complex system that has broken down. Many little things come together at the same time to produce failure.

Whereas Level 1 failures can be blamed on one individual, Level 2 errors usually involve a combination of internal and external causes. Examples of Level 2 failures include the problems of the supply chain during COVID, The Torrey Canyon ship hitting a reef at full speed and spilling 13 million gallons of oil.

Elizabeth Findell and Sadie Gurman (2023) reported on a 600-page United States Department of Justice report accounting for the failure of nearly 400 Texas law enforcement officers to rapidly intervene in a 2022 tragedy in which one gunman killed 19 fourth-graders and two teachers. This was a classic Level 2 failure: The first officers at the scene failed to treat the situation as an active shooting incident. Police later compromised the crime scene by refusing to cooperate with the FBI.

Edmondson says there may be some emotional relief when the company fires the leader of the group that committed Level 2 failure. Such firings focus on individuals and do not focus on systemic problems within the organization.

Board directors should retain the services of outside experts to investigate the systems issues involved in Level 2 failure. The experts should report to the board and not to the CEO. Outside experts look for the small warning signs that preceded the failure. Such warning signs are usually missed, ignored, or downplayed.

At a supervisory level, reducing the probability of Level 2 failures involves focusing on team training in simulated crises. Supervisors should never assume things will go well. They should create failure scenarios and then have the team rehearse responses repeatedly. Such repetition increases the probability that team responses in real situations will be automatic.

Level 1.5 Failure

Edmondson does not discuss this in her book, The Right Kind of Wrong: The Science of Failing Well, but we frequently see Level 1.5 failures in our consulting practice. A Level 1.5 failure is a Level 2 failure, and everybody recognizes that there is something fundamentally wrong with the system. However, they do not know how to address the problem from a system perspective.

Their intervention pretends the problem is a Level 1 failure. A classic example is focusing on employees on your team who are constantly late for work. It is addressed as a behavioral problem. But it could also be a symptom of a corporate culture that arbitrarily requires a time-sheet mentality unrelated to the business needs. It could be that the company fails to appreciate the logistical difficulty of taking children to school and then driving off to work.

Passive-aggressive team members may be a symptom of too much work and too few resources to manage the work.

Level 3 Failure

Edmondson calls Level 3 failure “intelligent failure.” Intelligent failures involve problems associated with grappling with novel situations.

According to Thomas Edison’s records, he failed 2,774 times to find a light bulb filament that would glow in a vacuum when electricity passed through. Treating this type of exploratory failure the same way you treat Level 1 or Level 2 failure defeats the purpose of research. Such failures should be celebrated.

Level 3 failures often occur outside the laboratory. An appropriate Level 3 failure is an attempt to climb Mount Washington in New Hampshire only to discover that you are not physically up for the task. It could involve baking a pie for the first time with results that are not tasty.

A primary care physician experiments with a new medication, a sales professional tries a novel approach to bring prospects, and a store manager changes the usual assortment on the shelves.

Edmondson argues that Level 3 failures be encouraged since they are the only way to achieve Level 3 success. Create a safe environment for employees to try novel approaches. One company established Failure Fridays. They allow colleagues opportunities to safely share what did not go well and what was learned.

Boards should encourage CEOs to report Level 3 failures as a way of encouraging a corporate culture of innovation. We recommend one hour a year be devoted to formal presentations of Level 3 failures within and outside the research and development function.

Conclusion

Failure is bad and success is good is a framework for failure. Not all failures are alike. There are four types of failures. They need to be managed differently. And the board needs to play its part in failure management.

Steve Jobs is the symbol of 20th-century entrepreneurial success. When he gave a commencement address at Stanford University, however, he chose to discuss the importance of failure in his development as a leader.