Learning from mistakes: when getting it wrong improves strategy

In this Italian-language video, walking along the Arno, I say that there is no such thing as a perfect strategy, and that mistakes, if they are read with a clear head, can lead to a step forward. Here I describe a study by Amy Edmondson, of Harvard Business School, that found more errors precisely where the work was best led, and I explain why not all mistakes deserve the same leniency.

Video transcript (translated from Italian)

Walking along a river, you understand that nothing stays the same.

The water flows, changes shape, adapts to every bend, but it keeps on flowing, always finding its way.

Every strategy should be like that: alive, changeable, able to adapt to time and circumstances.

There is no such thing as a perfect strategy.

Every company, sooner or later, makes mistakes, and it is precisely from those mistakes that an opportunity to take a step forward can arise.

Mistakes, if they are read with a clear head, are the moment when strategy renews itself, when something stronger is built, something more consistent with reality.

That is what a business strategist is for: to give shape to change, to correct the course while everything is moving, to turn the unexpected into opportunities.

To do that, though, knowing the numbers or the techniques is not enough.

It takes a real strategic culture, a vision able to bring together thought and action, intuition and method.

Because strategy is never static: it is like the water of the river. It moves, it adapts, and that is exactly how it goes on existing.

Imperfection is not a flaw to be corrected, but proof that something is changing, and that the change, if it is guided with culture and vision, can become the highest form of growth.

A mistake helps strategy only if it reaches the people who decide

Every company makes mistakes, and the people who run it know it. The difference between one company and another lies in how many mistakes come to the surface, and how quickly. A wrong order that nobody reports, or a lost customer nobody mentions in meetings, is a mistake the strategy will not see, and so cannot learn anything from.

In the video I say that mistakes, if they are read with a clear head, are the moment when strategy renews itself. For them to be read, someone must first have talked about them.

A mistake renews strategy only if the company sees it in time, and the company sees it only if the person who made it can say so without fear of paying for it alone.

In the best-led units, more errors were reported

In 1996, in the Journal of Applied Behavioral Science, Amy Edmondson published a study of eight units in two teaching hospitals. She wanted to find out whether better-led teams made fewer medication errors. The result went the opposite way: the units with stronger leadership, better relationships among their members and outcomes perceived as better had higher detected error rates. Reporting rates could also differ by as much as tenfold from one unit to another.

The explanation Edmondson gave has more to do with speaking up than with the work itself. Where there was a climate of openness, it was easier to report an error and discuss it, and where people feared the reaction of those in charge, errors stayed hidden. “The good teams, according to this interpretation,” she said years later, “don’t make more mistakes, they report more.”

In a study of units in two hospitals, published in 1996, Amy Edmondson found that the best-led units recorded more errors, and attributed this to a greater willingness to report them.

The study has limits worth stating: eight units, outcomes that were perceived rather than measured, and no way of separating the errors made from the errors reported. From there, though, Edmondson built the idea of psychological safety. In 1999 she defined it as a belief, shared by the members of a team, that the team is safe for taking interpersonal risks: saying that something is wrong, asking questions, admitting you do not know.

Not all mistakes deserve the same leniency

In the video I say that imperfection is proof that something is changing. In her book on failure, Right Kind of Wrong (Atria Books, 2023), Edmondson helps clarify which mistakes we are talking about. A basic failure happens in familiar territory, where people knew how to do the job and got one step wrong. A complex failure comes from several causes adding up, likewise in familiar situations. An intelligent failure happens in new territory, in pursuit of a goal and starting from a hypothesis, and it is only as large as it needs to be to learn something.

Not all mistakes are equal: those made in familiar territory should be reduced as far as possible, and only those made in new territory, small and born of an experiment, should be allowed for.

Tolerating basic failures in the name of experimentation is carelessness. Punishing intelligent ones pushes a company to stop trying new things, and leaves it with a strategy that stops correcting itself. It is the difference between direction and path that I talk about when people ask me whether the strategy changes during the work: the direction stays, and the path is corrected precisely through intelligent failures. It is also what allows a strategy to take shape along the way.

High standards and safety go together

There is a serious objection, and the people who run a company are right to raise it: if no one answers for their mistakes, no one will make an effort to avoid them. Edmondson addresses it directly. Psychological safety, she wrote in the Harvard Business Review in 2008, is “not about being nice,” nor is it “about lowering performance standards.” It is independent of each person’s accountability: healthy organizations foster both, setting high goals while openly acknowledging what is not yet known.

One figure helps show where the problem lies. When she asks executives how many of the failures in their organizations truly deserve blame, Edmondson gets estimates of between 2 and 5 percent. When she asks how many are treated as blameworthy, the answers rise to 70 to 90 percent (Harvard Business Review, 2011). These are answers gathered by talking with them, not a measurement, but they show how often a mistake is treated as someone’s fault even when it is not.

How a company reacts to a mistake is decided first of all by the people who run it, and it is one of the clearest signs of company culture, the kind that is not written down anywhere. That is why it is part of the work on organization and culture, together with the criteria of a business strategy clear enough to say which risks are worth taking.

What to do and what not to do

• Respond to the first report of a mistake in a way that makes sure there will be a second.
• When something goes wrong, distinguish a mistake made in familiar territory from one made in new territory.
• Keep goals high and say so, and make it just as clear that people can say when something is wrong.
• Do not reward a department that never makes mistakes without asking whether it makes them in silence.
• Do not call a mistake that could have been prevented an experiment.

Sources

• Amy C. Edmondson, “Learning from Mistakes Is Easier Said Than Done: Group and Organizational Influences on the Detection and Correction of Human Error,” Journal of Applied Behavioral Science, 32 (1), 1996, pp. 5–28. https://doi.org/10.1177/0021886396321001
• Amy C. Edmondson, “Psychological Safety and Learning Behavior in Work Teams,” Administrative Science Quarterly, 44 (2), 1999, pp. 350–383 (the definition on p. 354). https://doi.org/10.2307/2666999
• Amy C. Edmondson, “The Competitive Imperative of Learning,” Harvard Business Review, July–August 2008. https://hbr.org/2008/07/the-competitive-imperative-of-learning
• Amy C. Edmondson, “Strategies for Learning from Failure,” Harvard Business Review, April 2011. https://hbr.org/2011/04/strategies-for-learning-from-failure
• Amy C. Edmondson, Right Kind of Wrong: The Science of Failing Well, Atria Books, 2023 (Italian edition: Il giusto errore. La scienza di fallire bene, translated by Marianna Grimaldi, Egea, 2024).
• “Are Great Teams Less Productive?”, interview with Amy C. Edmondson, HBS Working Knowledge, April 23, 2007. https://www.library.hbs.edu/working-knowledge/are-great-teams-less-productive