Like many people who study behavior change and innovation, I often advocate that people analyze their failures and learn from them in order to improve for the future.
People trying to improve their behavior or to create new viable ideas will have to face attempts that didn’t work. In fact, when speaking to companies, I suggest that they help to remove the stigma from projects that didn’t work by celebrating each year’s most spectacular failure.
This suggestion is typically greeted with a laugh.
But I'm serious. I think that companies should share their stories of failure. Most people and most companies, however, are not eager to share their failure stories.
This point was demonstrated in a 2020 paper by Lauren Eskreis-Winkler and Ayelet Fishbach in Organizational Behavior and Human Decision Processes. They developed a series of studies to explore reasons why people fail to share failures.
In one set of studies, participants played a simple game in which they had to pick one of three boxes. They were told that one box had a 1 penny loss, one box had a 20-cent gain, and a third box had an 80-cent gain. They played the game twice, each time picking a box. The game was set up so that in one game, they picked the box that lost them a penny. In the second game, they picked the box that won them 20 cents.
Then, they were told that they could share the results of one of their games with another player with the aim of helping that player do as well as possible. This setup puts sharing a failure at odds with the best strategy. If the player shares their worst score, then they give the other player the best help, because they have shown their partner the location of the only box containing a loss. So, at worst, the other player will get 20 cents. But, if they share their better score, the other player does not know where the loss is.
Even though the optimal strategy is to share their “failure,” about half the participants in the studies using this procedure shared the 20-cent win rather than the 1-cent loss. This was true, even when participants were given an incentive to help the other person as much as possible.
Why does this happen?
Part of the reason participants share their success is that they may think it reflects well on them. But subsequent studies in this paper suggest that participants don’t really understand how much can be learned from failures. For example, in one study, participants were told they could get a hint in the game in which they could be told the location of the 1-cent loss or the 20-cent gain. About a third of the participants selected the 20-cent gain as the hint, suggesting that they did not realize that finding the location of the 1-cent loss would enable them to avoid losing anything in the game.
A similar pattern emerged in another set of studies in which participants answered quiz questions by selecting one of two answers. On some trials, participants were told whether they were correct and on others they were given no feedback. Participants preferred to share questions with others on which they got no feedback rather than sharing wrong answers, even though sharing the wrong answer would enable the other person to get the question correct. In a version of this study in which participants were reminded that sharing their wrong answers would ensure that the other person would be correct, participants correctly shared their wrong answers.
These studies suggest that people may avoid sharing failures because they don’t understand the value that sharing a failure may have to others. Of course, these studies use a contrived laboratory situation. What happens in situations closer to the real world?
In another study in this paper, participants were school teachers who wrote about successes and failures they experienced while teaching. They were asked to state whether they would prefer to share a success or a failure. They also rated whether they thought they had learned anything from the experience. As in the lab studies, the teachers were more likely to share their successes than their failures. They also felt that there was more to learn from the successes than from the failures.
Overall, then, there is a consistent pattern both in laboratory studies and in field studies for people to share their successes more often than their failures. This behavior flies in the face of advice given by people who study behavior change and innovation that failures carry a lot of information about how to improve performance. However, in both sets of studies, participants did not always see the value of sharing failures.
As a result, if we want people to share more of their experiences in which they failed to achieve a goal, we also need to spend more time teaching people both how to learn from their own failures as well how other people can learn from what they did wrong.
Eskreis-Winkler, L. & Fishbach, A. (2020). Hidden failures. Organizational Behavior and Human Decision Processes, 157, 57-67.