Intelligent failure is the right kind of wrong: Amy C Edmondson

Q. Why do you think it’s imperative to reframe our thinking now?

The need for fast learning from failure is most critical in times of uncertainty and upheaval, in part because failures are more likely. Second, while encouraging people to work hard to avoid failures may help them focus, it’s still critical to encourage experiments and welcome the intelligent failures they bring; this is essential to progress in any industry. Third, leaders need to recognise that the most likely outcome of a prohibition on failure isn’t perfection but rather being left in the dark—that is, not hearing about the failures that do occur.

From the small setbacks we experience in our day-to-day lives to the unsuccessful attempts to save lives in the early days of open-heart surgery (on patients who had no other hope for survival), failures are an unavoidable part of progress. This is true for all the vital institutions that shape society – as well as in our personal lives. This is why it’s so important—and ultimately so rewarding—to master the science of failure.

Q. Why is it important to distinguish between failure types?

Not all failures are equal. Basic failures are preventable ones in familiar territory; they’re caused by deviations from specified standards or knowledge. Complex failures are those that arise from unique combinations of factors, also in, at least, partly familiar territory. Both basic and complex failures are theoretically, and, in most cases, practically preventable. In contrast, intelligent failures are those that occur quickly and on a small scale, in new territory, thus providing valuable information. Learning to recognise and learn from each of the three failure types is a lifelong process—and one that all organisations must work hard to nurture. Excellence today means learning to fail well: That is, to prevent basic failures as often as possible, anticipate and mitigate complex ones, and cultivating an appetite for more frequent intelligent failures.

Covid-19 had it all: Basic failure, complex failure, and ultimately even intelligent failure. Basic failures happened every time a leader ignored basic health guidelines and it led to unnecessary infections and even deaths. Complex failures happened with the breakdown of global supply chains, due to labour shortages, weather challenges, and much more. Last, the ultrafast development of a Covid vaccine involved smart, iterative experimentation in scientific laboratories and well managed clinical trials, to arrive at a successful vaccine in record time.

Also read: Is fear of failure stopping you?

Q. What’s the right kind of wrong?

I see intelligent failure is the “right kind of wrong”. To be intelligent, a failure must take place in new territory, in pursuit of a valued goal, with adequate preparation and risk. Intelligent failures are essentially the experiments that didn’t pan out: The hypotheses that didn’t turn out as hoped. They are essential to the development of new knowledge.

Embracing intelligent failure is a requirement for inventors, scientists, celebrity chefs, and company innovation labs alike. Such elite failure practitioners have much in common—they are driven by curiosity, they experiment fearlessly; in short, they make friends with failure. Emulating them can also help anyone live fuller, more adventuresome lives.

Q. What goes into the making of a healthy failure culture?

A healthy failure culture rewards intelligent failure. It rewards people for undertaking smart experiments. Some of those experiments succeed; others fail. Without experimentation, there can be no innovation. Without innovation, no organisation can survive over the long term. But ensuring that there are vaguely negative consequences for not trying can make a healthy failure culture even more powerful. When I visited the Moonshot factory (called X) at Google’s beautiful offices in 2019, Astro Teller told the assembled group of employees—in response to a question about layoffs leading to fear of failure—that neither he nor anyone else could promise that layoffs would never happen. But if layoffs were needed, he said, the first to go would be people who had never failed. Context is critical to interpreting that statement. In an ambitious innovation arm of any company, intelligent failure is vital to progress.

Tata Group similarly launched their Dare to Try Award to celebrate audacious attempts to innovate that failed. Winners included an engineering team at Tata that developed an innovative new transmission that was too expensive to be implemented, and another team that created safe and effective plastic car doors that met with consumer mistrust. NASA, in the aftermath of the tragic Columbia shuttle failure, instituted the Lean Forward, Fail Smart Award, to shift its culture to encourage people to speak up quickly with both ideas and concerns.

Q. How crucial is psychological safety in this context?

Psychological safety is the antidote to the interpersonal fear that prevents us from failing well. My research has shown that psychologically safe environments help teams avoid preventable failures. They also help them pursue intelligent ones. Psychological safety helps people do and say the things that allow them to learn and make progress. This interpersonal climate factor turns out to be crucial in predicting team performance in challenging environments, ranging from those in leading academic medical centres to Fortune 500 companies to your family.

Toyota prevents basic failures by creating an environment of psychological safety. Factory workers are invited to pull the Andon Cord when they suspect a potential error on any vehicle. Its symbolism—we want to hear from you, and we especially want to hear about problems so we can make things better—captures the overarching ethos of Toyota’s organisational culture.

Q. What’s system awareness? How can organisations benefit from it?

System awareness—especially understanding how systems can produce unwanted failures—is a crucial skill in the science of failing well. Practicing systems thinking starts with consciously expanding your lens from its natural preference for here and now to include elsewhere and later. Two simple questions can help: 1. Who and what else will be affected by this decision or action? 2. What additional consequences might this decision or action cause in the future?

Equally important is the opportunity to design systems thoughtfully to achieve particular goals. In one of the most provocative elements of 3M’s system— provocative, at least, in an era in which companies prioritised efficiency— engineers were allowed to spend 15 percent of their paid time pursuing crazy ideas that might turn out to be failures. Later adopted by Silicon Valley companies such as Google and IDEO, the policy reflected an understanding that paying scientists to experiment would produce a lot of failures along with an occasional breathtaking success. The economics work—so long as you’re patient. That is, so long as you expand the boundary of the system to include the future profitability of the company, not just the present.

Q. Are corporate failure parties and failure resumés effective ways of building a healthy failure culture?

Although these have become popular, much of the discussion in books, articles, and podcasts is simple and superficial—more rhetoric than reality. For instance, it’s clear that no company should celebrate a plant manager whose automobile assembly line fails fast and often. Ditto for today’s heart surgeons. No wonder we are confused. Fortunately, this confusion can be reduced by understanding the three types of failure, and how differences in context matter. When the context is high stakes, with moderate variability, such as passenger air travel, fast failure is not good advice. When the context is low stakes and highly uncertain, such as in scientific laboratories, experimentation is the only way to make progress.

Q. Is “fail fast, fail often” a healthy mantra?

The failure craze—the “fail fast, fail often” culture that wants us to embrace failure seemingly indiscriminately—takes inspiration from the intelligent failures that are essential to innovation. But it glosses over the varied failure landscape, which also includes basic and complex failures. Some failures are bad, not in the sense of immoral but in the sense of wasteful. Whether tragic (a lost life) or silly (spilled milk), waste can be reduced through the diligent application of good failure-prevention practices. Basic failures are the most preventable of the three types and excellent companies strive to prevent as many basic failures as they can.