Artikels

Safety-I vs. Safety-II: how do you evolve toward a resilient safety culture?

For decades, safety was defined as the absence of incidents.

Fewer accidents, fewer deviations, fewer errors.

That is the classic Safety-I mindset: safety as “as few things as possible going wrong.”

For predictable risks, this model works. Technical hazards can be controlled through rules, procedures, and barriers. Incident investigations provide valuable insights. But modern organizations are complex systems. People work alongside technology, under time pressure, in changing conditions. In such environments, safety does not emerge from control alone, but from adaptation.

That is where the perspective shifts to Safety-II.

Safety-II defines safety as “as many things as possible going right.” It sounds simple, yet the shift is fundamental. The premise is that organizations succeed every day in carrying out complex activities. That success is not accidental. It results from professional expertise, experience, collaboration, and small adjustments when reality deviates from the plan.

Safety-II does not replace Safety-I. It enriches it.

Incident analysis remains essential. But if you only look at failures, you learn from exceptions. If you also study successes, you learn from the system’s normal functioning. And that is where the greatest learning potential lies.

How do you develop a Safety-II mindset?

Cultural change does not emerge from an inspiring keynote or a poster in the hallway.

It requires a structured trajectory.
Below is a practical growth path.

Strategic anchoring through leadership
- Organize a focused executive briefing on Safety-II and resilience engineering.
- Explicitly link it to business impact: operational continuity, reliability, reputation, stability.
- Clearly articulate why the organization is making this shift. Without strategic commitment, it remains a theoretical HR initiative.
Build knowledge and a shared language
- Train safety professionals and operational management in concepts such as work-as-imagined versus work-as-done.
- Introduce human factors and performance variability: variation is not only a source of risk, but also a source of success.
- Develop psychological safety to enable open reporting and honest dialogue.
Start with a pilot project
- Select one department or critical activity.
- Systematically analyze why it usually goes well.
- Identify buffers, informal expertise, and adaptations that make success possible.
  A pilot reduces resistance and provides tangible examples.
Integrate “Learning from Success” into existing structures
- Include success analyses in toolbox meetings or operational reviews.
- Explicitly ask: why did this go well?
- Capture lessons learned and share them across teams.
Rethink KPIs and dashboards – some examples:
- Number of “Learning from Success” analyses per quarter
  Measures whether the organization actively learns from normal work, not only from incidents. This measures whether the organization actively learns from normal work. It shows whether success is made visible and openly discussed.
- Number of improvement initiatives initiated from the working floor.
  This indicates the extent to which employees share deviations, optimizations, and smart adaptations. An increasing figure often reflects psychological safety and strong engagement. It is a proxy for adaptive capacity.
- Resilience capacity (buffer or redundancy index)
  For example: the percentage of critical processes with backup capacity or multi-skilled employees. This does not measure the past (what went wrong), but the readiness to absorb variability. It provides an indication of robustness before something goes wrong.

What you measure influences behavior. Different metrics create different conversations..

Build psychological safety.
- Encourage openness about how work is actually carried out.
- Avoid a punitive reaction to deviations from procedures.
- Encourage leaders to model behavior that prioritizes curiosity over blame.
Anchor in governance
- Integrate Safety-II into health and safety committees, project reviews, and management meetings.
- Make it normal to analyze success, not only incidents.
- Connect the approach to strategic objectives and long-term planning.

The core idea is that Safety-II should not become an additional layer on top of existing structures, but a different way of looking at performance.

Conclusion

The evolution toward a Safety-II mindset is neither a campaign nor a certification exercise. It is a fundamental shift in perspective. It is no longer only about asking how to prevent errors, but also how success emerges and how it can be made reproducible.

Organizations that make this shift strengthen their resilience. They better understand how their systems actually function. They build adaptive capacity. In a world where complexity is the norm, that becomes a strategic advantage.

Would you like to learn more abtou this topic? We recommand the following literature:

Hollnagel, E. (2014). Safety-I and Safety-II: The Past and Future of Safety Management.
Hollnagel, E., Woods, D. & Leveson, N. (eds.). Resilience Engineering: Concepts and Precepts.
Dekker, S. (2014). The Field Guide to Understanding Human Error.
Ham, H. (2021). Safety-II and Resilience Engineering in Practice.
Scanlon, M. & Jacobson, J. (2025). Safety-I, Safety-II and New Views on Safety. PSNet.
No, H. & Cha, S. (2025). Integrated Framework for Implementing Safety-I and Safety-II.

Michaël Marbais

Entrepreneur and Level 1 Prevention advisor.
As managing partner of Kingsm3n, S.Godart and Ken Do It, he helps companies turn safety, security, leadership, and people-centred policies into concrete action – in every routine and every crisis.