Beyond the Safety I and Safety II Debate: Why the Best Programs Use Both

Drawing on insights from a Solutions Review Solution Spotlight with Intelex, the Solutions Review editorial team explains why pitting Safety I and Safety II programs against each other is the wrong strategy. The better option is to use both.

The safety profession has a tribalism problem. On one side, practitioners defend the rigor and structure of traditional compliance-based safety management. On the other hand, advocates for newer, learning-oriented approaches argue that rules and audits alone cannot produce truly safe organizations. Both sides have compelling arguments. Both sides are also, in important ways, right.

That was the starting point for a recent Solutions Review Solution Spotlight conversation featuring Scott Gaddis, VP and Global Practice Leader at Intelex, and James Pomeroy, a global health, safety, and environment leader. The discussion explained why framing Safety I and Safety II as competing choices is the wrong question, and that organizations still asking which side to pick are missing the real work.

A Blended Approach: What Mature Programs Actually Look Like

The language of Safety I and Safety II doesn’t help the conversation. Calling one framework the “old view” and the other the “new view” is, as Pomeroy noted, a marketing move. It creates the impression of a clean generational break when the practical reality is far messier. Organizations cannot simply drop one model and adopt the other. Those who have tried to move too quickly toward Safety II without adequate groundwork have often found themselves ahead of both their leadership and their regulators.

What Safety I Gets Right and Where It Falls Short

Safety I is the foundation on which most organizations are built. It focuses on identifying and managing hazards, enforcing standards through compliance, conducting root cause analysis after incidents, and using hierarchical controls to eliminate or reduce risk. Done well, it produces structured, disciplined programs with genuine accountability. Most regulatory frameworks, including OSHA requirements, ISO 45001, and ANSI Z10, are rooted in Safety I principles.

The limitations are also real. A program that measures success exclusively through lagging indicators, injury counts, audit scores, and corrective action completion rates will tend to underreport near-misses, over-focus on minor compliance issues, and miss the systemic conditions that create serious harm. There is also a risk of what practitioners call the normalization of deviance, in which small deviations from standards accumulate quietly over time until a significant failure occurs. Compliance-based programs, by design, are better at catching visible violations than surfacing the invisible drift that precedes most major incidents.

The workforce dimension adds further pressure. Younger workers entering the trades and manufacturing environments expect participation, not just instruction. They are less receptive to top-down rule delivery and more likely to disengage from programs that treat them as subjects rather than contributors. A purely Safety I environment can struggle to build the trust and engagement that frontline safety actually requires.

What Safety II Adds and Why It Is Harder Than It Looks

Safety II emerged from a recognition that most work, most of the time, goes right. Rather than focusing exclusively on what failed, Safety II asks why things succeed. Learning teams, frontline participation, and the study of normal work become tools for surfacing the adaptations, informal practices, and contextual expertise that experienced workers develop over time but are rarely asked about.

The benefits are substantial. Frontline workers often possess genuine expertise that safety professionals sitting in corporate offices cannot replicate. When that expertise is invited into the safety process, programs gain insight they would otherwise miss entirely. Safety I views variance from procedure as a deviation to be corrected, whereas Safety II examines it as potential information. Sometimes, the deviation reveals a better way. Sometimes, it reveals a procedural gap. Either way, the conversation is more productive than a disciplinary response.

The practical challenges are equally real. Safety II requires trust before it produces results, and trust takes time to build. Learning teams require facilitation, time, and a genuine willingness to hear uncomfortable feedback. Metrics are harder to define because engagement and learning are less countable than injury rates. And leadership alignment is genuinely difficult, particularly for senior leaders who are distant from frontline operations and accustomed to managing safety through dashboards.

Why Choosing One Approach Creates New Problems

Staying entirely in Safety I produces programs that are structurally sound but culturally brittle. They can achieve compliance goals while failing to address the conditions that lead to serious incidents. Staying entirely in Safety II creates different risks. Without sufficient structure, programs can drift. Consistency becomes harder to maintain, regulatory alignment weakens, and the absence of clear accountability can allow standards to erode, becoming visible only after something goes wrong.

The Bradley Curve, which maps organizational safety maturity from reactive through dependent, independent, and interdependent stages, is useful here. The transition between the dependent and independent stages roughly corresponds to the point at which Safety I alone begins to reach its ceiling. Moving toward genuine interdependence requires the collaborative, participative elements that Safety II emphasizes. The curve does not suggest abandoning structure. It suggests adding capability.

How to Assess Where Your Organization Currently Sits

A useful starting point is examining how procedures are created, enforced, and reviewed. In a Safety I-dominant organization, procedures are typically written by safety professionals and handed down for compliance. Deviations are treated as violations. In a Safety II-oriented organization, procedures are developed through worker involvement, treated as guidance rather than absolute prescription, and revisited when people struggle to follow them. Neither extreme is ideal. The question is where your organization sits on that spectrum and whether that position reflects a deliberate choice.

Perception surveys and leadership interviews provide another diagnostic layer. The questions matter: Do workers feel safe raising concerns? Are near-misses reported and discussed, or quietly absorbed? Do leaders ask what went right as well as what went wrong? Are success stories shared with the same energy as incident investigations?

The measurement challenge is real but not insurmountable. Alongside traditional lagging indicators, organizations can track leading indicators, including participation rates in learning events, the quality of near-miss reporting, response times to worker-raised concerns, and perception survey results over time. The combination gives a more complete picture than injury statistics alone.

Workplace fatality rates in several countries, including the US, have stopped declining and in some cases are rising. That pattern suggests that doing more of what has always been done is not sufficient. The blended approach is not a theoretical preference. It is a practical response to evidence that compliance alone has reached its ceiling.

FAQ: A Blended Approach to Safety I and Safety II

What is the difference between Safety I and Safety II? Safety I focuses on preventing failures through compliance, hazard control, and root cause analysis. Safety II focuses on learning from normal work, engaging frontline expertise, and building adaptive capacity. Both address real aspects of organizational safety.

Can an organization use both Safety I and Safety II? Yes, and most high-performing programs do. The practical goal is knowing when to apply structure and accountability versus when to apply learning and participation, rather than choosing one exclusively.

How do you measure Safety II outcomes? Useful leading indicators include frequency and quality of learning team events, near-miss reporting rates, worker participation in safety processes, and periodic perception surveys. These complement rather than replace traditional lagging indicators.

Where do organizations typically start on the Safety I / Safety II spectrum? Most begin with Safety I. Moving toward a blended approach usually requires first building trust and psychological safety through consistent follow-through on worker input, then gradually introducing structured learning practices.

How do you get leadership buy-in for Safety II? Start by connecting Safety II practices to existing business priorities: workforce engagement, productivity, retention, and ESG performance. Senior leaders respond more readily when new approaches are framed in terms of outcomes they already care about.