Strategies for Learning from Failure

11/26/2015 – I am taking a course on entrepreneurial behaviour and practices and one of our key seminars has been entirely focussed on failure – imagine 8 days to explore all facets of and relationships to failure, with people from diverse backgrounds, organizations and cultures from throughout the world! I have loved it as it has helped me uncover and explore my own definitions and fears about failure (lots) and exposed me to new ways of thinking about and embracing failure as part of strong and innovative practice. One of the articles we reviewed is by Amy Edmondson, Novartis Professor of Leadership and Management at the Harvard Business School, entitled, Strategies for Learning From Failure (April 2011). I thought it was particularly beneficial and relevant to our sector. She noted that, despite being told that ‘innovation and failure is good’ many of us have been conditioned since childhood to avoid making a mistake and failing at all costs. This sets up a dynamic tension between what we are told is necessary in the workplace (e.g., think outside the box, innovate, test out new ideas, fail, learn) and what we have been told for most of our lives (e.g., don’t mess up).

She offers a more nuanced understanding of failure, that lines up well with the 2020 teachings on the Cynefin framework. With problems that are simple or complicated – what she calls predictable or knowable problems – failure should be preventable and avoidable and when it does show up it could be due to deviant behaviour (someone doesn’t want to follow accepted practices) but more likely due to inattention, lack of ability or training, process inadequacy or a task challenge (e.g. the task is simply too much to expect someone or a group to complete with the available resources). Leaders and managers need to act to prevent these types of failures.

On the other hand, when we face complex challenges where there is a high degree of uncertainty and we have not encountered the situation before, failures are often unavoidable – “system failure is a perpetual risk”. Our response should be one of trying to prevent more significant or “consequential” failures by attending to small process failures: “To consider them bad is not just a misunderstanding of how complex systems work; it is counterproductive. Avoiding consequential failures means rapidly identifying and correcting small failures.” She strongly encourages leaders to create an environment in which people can comfortably raise concerns and identify the small errors that can be addressed before the whole system is compromised

Edmondson also describes a third type of failure that occurs within the complex domain: “intelligent failures at the frontier”. I love this idea of intelligent failures – where experimentation is necessary and the “answers aren’t knowable in advance because this exact situation hasn’t been encountered before”. Here she suggests small-scale tests to see what might work and continual iteration and improvement – NOT large scale pilot projects! Failure here is “praiseworthy” as it serves learning and development.

Her description of “the blame game” was particularly relevant to our sector where, due to the high risks associated with some failures there is often intense media and public scrutiny aimed at placing blame on individuals or organizations. She would suggest that this generates a widespread lack of “psychological safety” that shuts down innovation and experimentation due to fear of failure – even when the risks are minimal or the benefits of new approaches could be significant. The five elements that Edmondson suggests leaders should cultivate in order to contribute to a psychologically safe environment that can prevent failure where appropriate, and be intelligent and learn from failures in the complex domain, include:

  • Frame the work accurately – what kinds of failures can be expected and how they will be worked through helps to ‘detoxify’ failure;
  • Embrace messengers of bad news, questions, concerns or mistakes;
  • Acknowledge limits – “be open about what you don’t know. Mistakes you have made, and what you can’t get done alone will encourage others to do the same”;
  • Invite participation – “ask for observations and ideas and create opportunities for people to detect and analyze failures and promote intelligent experiments. Inviting participation helps to defuse resistance and defensiveness”;
  • Set boundaries and hold people accountable.

You can gain access to the full article here: and signing into the Harvard Business Review. You can view online or get up to five free articles to download. The 12-minute interview with Amy is also available at this site and offers a good summary.

You may want to consider the following questions about ‘failure’:

  • What is your attitude or approach towards failure?
  • Are all failures equally ‘good’ or ‘bad’?
  • Where is a greater openness to failure and the consequent learning, needed in your own practice?