Why punishing failure is killing innovation performance and trust

Here is something most people leaders already know in their bones but rarely say out loud: the organisations that speak most confidently about innovation are often the same ones where people are still afraid to make a difficult call.  Many organisations say they want innovation, experimentation and learning. Yet McKinsey describes one telecom company where a single difficult decision was pushed up to the CEO four separate times because people were afraid of making the wrong call.

That gap between rhetoric and reality carries a real business cost. Where people do not feel safe to speak up, test ideas and learn from mistakes organisations lose insight, slow down decision-making and make innovation harder than it needs to be. Harvard Business Review Analytic Services found that innovation leaders are more likely than innovation followers to favour an experimental approach, by 46% to 30%. McKinsey has also found that the most innovative companies are more than twice as likely to reward appropriate risk-taking. Yet the lived experience of failure in most workplaces has barely changed. Bold ideas get quietly shelved. Preventable mistakes go unreported. People self-censor because they have learned that being wrong carries a cost.. 

Ed Catmull, co-founder of Pixar, put it honestly: failure in hindsight is a teacher, failure in prospect is a threat. Where mistakes signal incompetence, that threat is career-defining and the result is invisible stagnation.

This article is about closing the gap through deliberate work across four dimensions: leaders modelling vulnerability, cultural practices where experimentation is a daily discipline, formal mechanisms that capture learning before it evaporates, and organisations helping individuals move through the emotional weight of getting things wrong. Because failure does not just show up in project dashboards. It shows up in the body, in identity, in the story people tell themselves at three in the morning. Ignoring that dimension is why so many culture programmes stall.

The four pillars of a safe-to-fail environment

Pillar 1: Leadership attitudes and modelling

Culture change begins at the top, and here it begins with something most senior leaders have been trained to avoid: vulnerability. When a leader openly acknowledges a misstep, not in a rehearsed way, but with genuine candour, they send a signal that reverberates through the organisation: failure is survivable, learning from it is expected, and nobody needs to pretend they have all the answers.

Amy Edmondson, who pioneered the concept of psychological safety, distils the leadership task into three practices: frame work as a learning opportunity, not a test of competence; actively invite participation; and respond to setbacks with forward-looking questions rather than blame. The shift from “How did this happen?” to “What have we learned?” sounds small. In practice it changes everything.

But telling leaders to “be more vulnerable” is like telling someone to “be more charismatic.” The instruction is clear, the execution is not. Most senior leaders built careers on competence and control; admitting uncertainty feels like a risk to the identity that got them promoted. Therefore, the development work must be specific. Give leaders facilitated spaces to practise sharing setbacks. Build “models learning from failure” into competency frameworks, 360s, and succession criteria. Coach for the emotional dimension; leaders experience the same flush of shame as everyone else, and programmes that acknowledge this produce leaders who are more credible, trusted and resilient.

The point is not to turn every executive into a storyteller of personal failure. It is to make honesty about mistakes so ordinary that teams stop bracing for blame and start focusing on what comes next.

Pillar 2: Cultural practices and norms

Leadership sets the tone, but culture is sustained by daily habits. Three practices stand out.

Designed experimentation. Most organisations treat experimentation the way they treat annual budgets – one carefully controlled attempt that had better deliver the right answer the first time. Design the perfect test, wait for a clean answer, move on. In practice, experiments routinely “fail” in discouraging ways: poor design, insufficient sample size, missing data, inconclusive results or not learning from prior failures. All of those apparent failures are rich with information. The real skill is reframing experimentation as incremental: each round of results sharpens the next question. Herbert Simon, the father of design thinking, said that complex systems can only be understood by building and observing them. That principle became foundational to design thinking's deliberately failure-friendly methodology. For leaders: build experimentation into team rhythms as a discipline, not an event. Celebrate the clarity that comes from a hypothesis proved wrong.  Design of experiments provides the statistical rigour; design thinking provides the human-centered mindset; together, they shift failure from a verdict into a feedback loop.

Structured debriefs and post-mortems. If experimentation generates learning, debriefs capture it. Google insists post-mortems must be blameless. The facilitator opens by stating that the review exists to understand systems, not assign fault. Questions shift from “Who deployed the broken code?” to “What process allowed this to reach production?” Blameless culture correlates directly with reliability improvements, because teams that fear punishment stop reporting honestly.

There is a deeper layer Western frameworks often miss. The Japanese practice of hansei, “to turn back and examine,” goes beyond what happened to interrogate the thinking behind it. Katie Anderson documents how Toyota leader Isao Yoshino practised hansei not as punishment but as a path to understanding; even successful projects were followed by reflection meetings. Blameless post-mortems provide structural safety; hansei provides depth. Together, they move teams to “What were we assuming, and why?” where the transformative insights live.

Storytelling. Sharing narratives of failure and recovery builds empathy, normalises risk-taking, and transmits cultural values more effectively than any policy document. The specificity matters: not “we encourage experimentation” but “here is the test we ran, here is what surprised us, and here is what we changed.” That candour is contagious.

Pillar 3: Corrective mechanisms and tools

Good intentions will only get an organisation so far. Without formal systems that capture the lessons of failure, even the most psychologically safe culture will repeat the same mistakes and people will notice. Root cause analysis, structured feedback loops, and near-miss reporting all serve the same purpose: a disciplined process for asking “what do we do differently?” rather than “whose fault was it?” The challenge is embedding them so deeply that they become reflexive rather than ceremonial. What matters is cadence and follow-through: short-cycle reviews tied to real decisions, with visible evidence that findings actually change how the team works.

Hald, Gillespie, and Reader’s review of 74 case studies of institutional failure draws a useful distinction between “causal culture,” practices that create problems, and “corrective culture,” an organisation’s capacity to fix them. The most dangerous state is when both are dysfunctional. For people leaders the takeaway is pointed: corrective mechanisms are not a nice-to-have. They are the difference between self-correction and sleepwalking into crisis. A 2025 study demonstrated this in practice: when companies implemented near-miss management systems, employees perceived them as a tangible sign of commitment to learning rather than blame. Near-miss reporting transforms potential failures from hidden liabilities into visible learning opportunities.

Pillar 4: The individual mindset shift

An organisation can get structures, systems and leadership signals right and still watch talented people play it safe because failure does not just live in project dashboards. It lives in the body: the tight chest before a presentation, the hot flush of shame when an initiative falls flat, the corrosive replay of “what was I thinking?” at three in the morning.

The roots run deep. From schooling onwards most people absorb a relentless message: wrong answers are bad, the safest path is the one already validated. That conditioning does not evaporate with a senior title. If anything, the stakes feel higher. A 2025 study found that psychological safety alone was not enough to drive innovation; people also needed to experience “thriving at work,” vitality and ongoing learning, before translating 

A call to action for people leaders

The evidence is not in question. The challenge is that building a safe-to-fail culture is genuinely hard, harder than most change programmes acknowledge, because it asks people to unlearn a lifetime of conditioning. From red pen on homework to performance ratings that conflate a failed experiment with a performance problem, the message has been consistent: getting it wrong is dangerous.

Undoing that conditioning happens through sustained, unglamorous work across all four dimensions: leaders who model honesty about their own stumbles, cultural practices that make experimentation a discipline, formal mechanisms that capture learning before it evaporates, and the deeply human work of helping individuals move through the emotional weight of getting things wrong. None of these pillars is sufficient on its own. Together, they build something no rhetoric can fake: an organisation where people genuinely believe it is safe to try, to stumble, and to come back smarter.

The returns are substantial, more innovation, greater resilience, stronger retention. But perhaps the most compelling reason is simpler. The organisations that will thrive are not those that avoid mistakes but those that recover fastest, learn most honestly and have people willing to say, “I have an idea and it might not work.”

That is a willingness culture, a safe-to-fail culture. And building it is work for organisational leaders.