How does your company deal with mistakes? If continuous learning from your employees, innovation and even breakthrough innovation are important to you, it is critical you embrace mistakes, at least as sources for learning and invention.

It is a cliché to say that we learn from our mistakes. Indeed, we tend say it more than we practice it. Consider learning in school. In most cultures, young children are taught correct answers to basic questions, such as maths equations, spelling, grammar and the like. Then they are taught to repeat these answers over and over again until the question and answer pairs are embedded in their young brains and those children can say six times six is 36 without even thinking about it. Moreover, the assumption is that if a child were at some point to start giving an incorrect answer, such as six times six is 38, that wrong answer would embed itself in her mind and she would have to work doubly hard to unlearn the wrong answer and then learn the correct answer. Hence the emphasis on memorising correct responses to such questions.

It’s compelling logic. But it is wrong. Research(¹,²) has shown the opposite to be true. Children actually learn better when they are put in situations that increase the likelihood that they will make errors. A number of experiments were designed to test this assumption. In each of these experiments, children were given information to learn. In one group, the children were given tasks in which they first had to attempt to retrieve on their own answers to difficult questions. In this group, most of the children did indeed get their initial answers wrong. Then, they were given the correct answers to the questions. It was found that the children in this group learned the information better than the children the group that was taught the information using the old-fashioned rote-learning method.

Don’t reach for Google

As a manager, the lesson to be learned is not to tell starter employees precisely how to do their tasks.

As an adult, there is a lesson to be learned here. If you believe you do not know the answer to a challenging problem, do not immediately Google the problem. Instead, attempt to work out the answer using your own knowledge, experience and assumptions. Then Google. By so doing, you are more likely to learn the correct solution to the problem. Moreover, the process of correcting yourself is likely to be enlightening as well. Why was your guess wrong? What assumptions did you make in the process of making your guess? Alternatively, you may discover that your solution is more effective than what you have found on Google. That’s even better!

As a manager, the lesson to be learned is not to tell starter employees precisely how to do their tasks. Rather, you should let them attempt to work out for themselves how to solve problems – and then, if they fail, give them instructions. This has a twofold benefit. Firstly, new employees will learn how to perform tasks better this way. Secondly, in attempting to come up with their own approach to solving tasks, they may actually come up with a better approach than your company is currently using. That is the first step in the path to process innovation! It goes without saying that when employees might harm themselves or others, this is not the best approach to take – unless of course you are using simulations or another method to prevent injury.

At a higher level of business, there often are not defined means of accomplishing tasks. Indeed, it is at this level that you want employees to demonstrate creative thinking in order to solve problems. Fortunately, if you have allowed starter employees to learn by making mistakes, experienced employees are likely to be both more knowledgeable and more creative in their thinking. This can only work to your company’s advantage!

Mistakes: an alternative to training

There is a story, most likely apocryphal, of a manager who launches a major project only to see it fail spectacularly. Rather than revolutionising the business, his company loses $20 million. The manager sends his letter of resignation to the CEO and is in his office gathering his things when his secretary tells the manager that the CEO would like to see him right now.

The manager goes into the CEO’s office and immediately launches into an apology: “I know I blew it with that project. I’ve sent you my letter of resignation. I am sorry.”

The CEO takes the letter of resignation, tears into pieces and says, “son, that $20 million was the most expensive on-the-job training in the history of this company. If you think I am going to let you leave after all you have learned, you must be out of your mind.”

By seeing what did not work, I was able to visualise models that and behaviours that I believed would work better.

And the truth is, we do learn from mistakes. I learned a great deal about organizational innovation by spending a number of years working in organizations that did everything they could to hinder innovation. By seeing what did not work, I was able to visualise models that and behaviours that I believed would work better. Then, first by looking at successful innovators as well as social-psychological research, I was able confirm some of my theories (and abandon others). Later, in my current work, I have been able to put these models to the test.

Simply looking at successful innovators would not have been enough to allow me to fully understand what works and what does not work in terms of organizational innovation. Looking at failed innovators taught me much more.

Not admitting mistakes is more costly

In organizations where mistakes are considered unacceptable, employees are understandably reluctant to admit it when things go wrong. As a result, when a project is failing, the employee in charge is often inclined to continue the project, rather than admit failure. This can be an expensive waste of resources. A hopeless project killed early will cost the company far less than a hopeless project that is allowed to eat resources for months in hopes a miracle will occur. Clearly, even in companies where mistakes are unwelcome, this is not a desirable approach!

This scenario demonstrates two problems with respect to the fear of mistakes. One is the reluctance of people to admit to a mistake, even when it is clear things are going wrong. The second is a reluctance to ask for help to solve a problem, out of fear that asking for help will flag the mistake. Hence, not only are employees reluctant to admit that a project is failing, they are equally reluctant to ask for help that could prevent, or even reverse a failure!

That is a shame. Not only is there much to learn from mistakes. They can also spark off breakthrough innovations.

Some mistakes lead to great inventions

Perhaps the most famous story of a great invention being discovered as a result of an accident or mistake is the discovery of Penicillin by Alexander Fleming³. He had been studying staphylococci , a kind of bacteria that can often cause illness in humans. Just before leaving on holiday, he stacked a number of cultures of staphylococci on his lab table and left them there (apparently, he was notoriously untidy). Upon his return, he observed that one culture was contaminated with a fungus, and that the colonies of staphylococci that had immediately surrounded it had been destroyed, while the colonies further away were normal. The fungus was identified as being of the Penicillium genus and hence the name of the first antibiotic and arguably one of the most important discoveries in medical history.

There are two “what-ifs” to consider here. Firstly, what if Mr. Fleming was working in a large company in which mistakes were not tolerated. Upon returning from holiday and seeing that his lab was a mess, or that something was not right with some of his cultures, his first reaction might have been promptly to clean and sterilise everything before his superiors discovered his mistake.

Alternatively, what if Mr. Fleming did not have the medical knowledge that he did actually have. Or what if he simply did not have an inquisitive mind. In either case, he might not have realised the powerful implications hidden in the Petri dish with the fungus.

When people make mistakes, when projects go wrong, when teams screw up, their errors should not be swept under the corporate carpet and forgotten.

However, the world is fortunate that Fleming was not employed in a bureaucratic organization with zero tolerance for mistakes and that he was a brilliant biologist with an inquisitive mind.

What your organization can learn from mistakes

There is no doubt about it, mistakes are proven learning exercises . So much so that managers should sometimes encourage mistakes, especially during the early learning phase of new employees. When people make mistakes, when projects go wrong, when teams screw up, their errors should not be swept under the corporate carpet and forgotten. Rather, they should be shared so that not only the people making the mistakes, but also their colleagues can learn from those mistakes.

Moreover, you never know. By sharing the events that led up to a mistake and the results, someone in your firm may observe an opportunity to profit. More than one breakthrough innovation has been discovered this way. And many more will be – perhaps some of them will be discovered by you and your colleagues.

By Jeffrey Baumgartner

About the author

Jeffrey Baumgartner is the author of the book, The Way of the Innovation Master; the author/editor of Report 103, a popular newsletter on creativity and innovation in business. He is currently developing and running workshops around the world on Anticonventional Thinking, a new approach to achieving goals through creativity.


1) “Unsuccessful retrieval attempts enhance subsequent learning.” (July 2009) by Kornell, Nate; Hays, Matthew Jensen; Bjork, Robert A.; Journal of Experimental Psychology: Learning, Memory, and Cognition, Vol 35(4), 989-998.

2) “The Pretesting Effect: Do Unsuccessful Retrieval Attempts Enhance Learning?” (2009) by Richland, Lindsey;  Kornell,Nate and Kao, Liche; Journal of Experimental Psychology: Applied,  Vol. 15, No. 3, 243–257

3) “Alexander Fleming” (2001-2010); Wikipedia;