A team trials a process change, it makes things worse for two days, and the retrospective conversation never quite happens. Nobody documents what went wrong, nor raises it in the next review. The experiment simply disappears, and with it, the learning that could have come from just that. Multiply that pattern across an organisation and you get teams that look busy with improvement activity while capability quietly stalls. This is a conditions problem, usually set by leaders. Tim Edwards, Senior Operations & Improvement Manager at LCS explores just this.
Google's Project Aristotle, a two-year study of 180 internal teams, found that psychological safety, whether people feel safe taking risks and being seen making mistakes in front of colleagues, was the single strongest differentiator between high-performing teams and the rest (Google re:Work). Continuous improvement depends on exactly that kind of risk-taking, which makes psychological safety a leadership responsibility, not a personality trait some teams happen to have.
Few meaningful improvements work on the first attempt, especially inside complex operational systems, and many organisations continue to reward certainty and immediate success far more consistently than they reward learning itself. Teams operating in that environment can be actively encouraged to innovate on paper while working inside a culture where an unsuccessful outcome is treated as a personal failure rather than useful information.
The result is predictable because people stop bringing forward the ideas most likely to teach the organisation something, because those are also the ideas most likely to fail visibly. Improvement activity narrows to safe, low-risk tweaks, and the capability that comes from genuine experimentation never gets the chance to develop.
Psychological safety is often misread as an absence of standards, a soft environment where nothing is challenged. The opposite is however closer to the truth. Teams that feel safe are more willing to surface problems honestly, challenge an approach that isn't working, and hold each other to account, precisely because doing so doesn't carry personal risk.
The distinction leaders need to hold is between the outcome of an experiment and the quality of the thinking behind it. A well-designed test that doesn't deliver the expected result isn't a failure of standards; a team that stops testing anything because it's afraid to be wrong is. Leaders who conflate the two, treating any unsuccessful attempt as a discipline issue, train their teams to stop trying rather than to try more carefully.
There's a second shift alongside psychological safety with that being what leadership itself requires is changing, and it's changing unevenly across levels. Practitioners working within operational teams still need strong technical foundations alongside the ability to solve problems collaboratively in live environments.
Leaders however who are responsible for enterprise-wide improvement increasingly need something broader: systems thinking, strategic alignment, cultural influence, and the ability to sustain capability across parts of the organisation they don't directly control. These are not the same skills that make someone an excellent practitioner, and treating them as interchangeable is a common reason capable technical leaders struggle once their remit becomes strategic. As organisations become more complex and adaptive, this behavioural and systemic dimension of improvement leadership only becomes more important, not less.
Psychological safety isn't built through a single announcement or a value poster placed on the wall, instead it is built through how a leader responds the first time someone tells them an experiment went wrong. Curiosity in that moment, rather than blame, teaches a team more about what's genuinely welcome than any policy document could.
The same applies to how leaders talk about their own decisions. A leader willing to say a past call didn't work out, and to explain what they learned from it, gives everyone else permission to do the same. These are small moments, repeated often, and they accumulate into a culture far more reliably than any single intervention.
LCS Tip: In your next review, ask your team a genuine “what didn't work this week, and what did we learn from it?” How the room responds will tell you more about your psychological safety than any survey.
Adaptive improvement capability doesn't happen by accident and instead requires leaders who actively create space for experimentation, respond to failure with curiosity rather than judgement, and recognise that their own capability needs to keep evolving alongside their teams'.
This is precisely the gap that structured leadership development and professional accreditation are designed to close, giving organisations a consistent standard for the judgement, influence, and systems thinking that strategic improvement leadership now demands, rather than leaving it to whoever happens to have developed those skills informally.
We explore the wider shift this reflects in our whitepaper, Building Adaptive Improvement Capability Beyond the Classroom. It builds on the themes covered in Continuous Improvement Training Is Only the Beginning and Building Continuous Improvement Capability Through Real Work.
If you're developing the next generation of improvement leaders in your organisation, find out more about LCS accreditation and membership and how the LCS Framework supports leadership capability alongside practitioner development.