Keynote Title: From Metrics to Meaning: Reconstructing Safety at Southern Company
“Our TRIFR is the highest it’s ever been in 20 years, yet we are safer than ever…”
Kim Greene, CEO Georgia Power at the Safety On The Edge Conference 2025
Abstract
This paper explores Southern Company’s ten-year transformation of its safety strategy—from a compliance-focused, metrics-driven approach toward a culture rooted in risk awareness, organizational learning, and human-centered leadership. Drawing on a speech by CEO Kim Greene, it examines the limitations of traditional lagging indicators such as Total Recordable Incident Rate (TRIR), and the role of leadership in enabling cultural change. The journey offers valuable insights for safety professionals, executives, and boards seeking to move beyond the illusion of safety toward more resilient and trustworthy systems of risk management.
Introduction
Across industries, Total Recordable Incident Rate (TRIR) has long served as a staple metric for safety performance. Lower numbers are widely interpreted as evidence of success; they adorn dashboards, determine bonuses, and often serve as the proxy for safety culture maturity. Yet in 2024, Southern Company—a major U.S. electric and gas utility—reported its highest TRIR in over two decades. Paradoxically, the organization simultaneously recorded its lowest ever number and severity of serious injuries.
This apparent contradiction forms the foundation of a powerful lesson: safety performance cannot be reduced to recordables alone. Instead, meaningful safety requires rethinking not only what we measure but how we lead, learn, and care.
The Fallacy of “Target Zero”
Southern Company’s safety journey began in earnest in the early 2000s, following a series of fatalities that triggered a company-wide reckoning. In 2003, the then-CEO declared that such outcomes reflected not a worker problem but a leadership failure. This led to the adoption of “Target Zero,” a strategy that prioritized eliminating all recordable injuries.
The focus on Target Zero was well-intentioned but, in hindsight, misguided. It aligned with behavior-based safety (BBS) principles that emphasized individual accountability and compliance. Workers who were injured were brought before executives to explain their errors. Awards and recognition were granted for record-free periods, further reinforcing an environment where reporting injuries jeopardized group morale and rewards.
TRIR improved steadily during this period, creating the impression of safety progress. However, serious injuries and fatalities remained stubbornly static—a pattern consistent with industry-wide findings challenging Heinrich’s Triangle and its assumptions about proportionality between minor and major events.
A Leadership Inflection Point
By 2014, a reversal in TRIR prompted deeper scrutiny. Kim Greene, then Chief Operating Officer, began questioning the company’s assumptions. Consultation with academic safety experts revealed a sobering assessment: while compliance efforts had improved, the organization lacked the systems, culture, and leadership frameworks necessary for true risk control.
A pivotal moment was Southern Company’s first safety culture survey using the Prisma Culture Assessment methodology. The results were striking. Workers perceived that leadership valued metrics over people. Safety systems were fragmented across business units. Risk awareness was low, especially among younger field staff. Injuries were routinely underreported to preserve reputational or financial standing. The cultural profile placed the organization between “rational” and “righteous”—far from the resilient culture it aspired to.
These findings exposed what many safety professionals now refer to as risk secrecy: an organizational condition where good numbers mask systemic vulnerabilities.
Reframing Success: From Compliance to Risk Awareness
The company’s leadership—supported by the Board—embarked on a fundamental realignment of its safety strategy. The decision was made to remove TRIR from incentive plans, despite internal resistance. In its place, new performance indicators were introduced, emphasizing:
• Serious Injury and Fatality (SIF) events
• Maturity of the Safety and Health Management System
• Critical risk exposure and control effectiveness
• Participation in learning processes (e.g., event reviews)
• Progress in safety culture maturity
This reframing marked a shift from counting failures to evaluating system resilience—an approach consistent with modern safety science, including Safety-II and human and organizational performance (HOP) principles.
Risk awareness also became a central pillar. High-risk activities—such as confined space entry, electrical work, and line restoration—were identified and prioritized for training and procedural redesign. The company embraced the hierarchy of controls, emphasizing elimination, engineering, and administrative protections over procedural compliance alone.
Building a Learning Culture
A key aspect of the transformation involved reshaping the company’s approach to incident analysis. Rather than interrogating injured workers, Southern Company adopted learning reviews based on openness, reflection, and systemic diagnosis. Over nearly a decade, the organization fostered psychological safety around event reporting—an essential but often elusive condition.
Leaders also embraced human fallibility as a design condition. Rather than expecting flawless behavior, the organization focused on building error-tolerant systems. Greene likened this to aviation, where checklists and procedural discipline are standard practice. In her words, “We never took off without a checklist. Why did our industry see checklists as weakness?”
Today, safety conversations at Southern Company involve engineers, frontline workers, and executives alike. The Safety and Health Management System now codifies expectations that transcend individual leaders and persist across operational transitions.
Outcomes: Reporting More, Hurting Less
The results of this transformation are telling. TRIR increased—largely due to improved reporting—but serious injuries declined to the lowest levels in company history. More importantly, when incidents occurred, they were less severe. Medical claims costs, pharmaceutical spend, and legal expenses also declined, reinforcing the business case for safety—a topic often left unspoken but increasingly relevant for executive buy-in.
Culturally, the company saw a marked shift. A second safety culture assessment, conducted nearly a decade after the first, revealed a dramatic rise in employee trust, system coherence, and organizational learning. Southern Company was placed in a new category of resilience not previously reached in its industry segment.
Lessons for the Field
Southern Company’s experience highlights several enduring insights for safety professionals and executive leaders:
1) Lagging metrics mislead. A low TRIR may reflect underreporting, not safety. As Greene stated, “You pay people on TRIR, and you’ll get a lot—but you won’t learn from it.”
2) Behavior is not the problem—systems are. Blame-based models inhibit learning. Humans are fallible; systems must account for that reality.
3) Risk awareness trumps rule-following. Frontline employees must understand the specific hazards of their work. Generic compliance is insufficient.
4) Psychological safety enables reporting. Learning depends on trust. Fear of punishment drives silence and hides risk.
5) Leadership courage is indispensable. Transformational change requires executives to stand firm—even in the face of resistance or political pressure.
Conclusion
The illusion of safety is persistent, seductive, and dangerous. Southern Company’s journey illustrates how even organizations with strong intentions and improving statistics can miss the mark when safety is reduced to numerical targets. True progress lies not in eliminating recordables but in enabling risk awareness, system resilience, and learning.
The challenge is not merely technical; it is moral and cultural. Safety must be championed not through metrics, but through meaning. Southern Company’s story proves that such transformation is not only possible—it is essential.