Behavioral Observations: A Double-Edged Sword for Workplace Safety?

Over the last 25 years I have seen numerous well-intended workplace health and safety initiatives come and go. Yet Behavioral Observations continue to be a mainstay across high-hazard industries.

In my experience, there is a great deal of variability in the intent, facilitation and outcomes associated with such behavioral approaches.  I have seen observations used in ways that enabled genuine workforce engagement and safety improvements, and often in ways that led to reduced trust, increased cynicism and low psychological safety.

In this piece I first outline what I see as the perceived benefits associated with behavioral observations, followed by potential downsides. I share these not merely to critique, but to help readers understand and mitigate the potential risks inherent to this approach. 

Perceived Benefits of Behavioral Observations

1.  Immediate Action & Tangibility: They provide a quick, simple, visible way to "do something" about safety, focusing on concrete actions managers and workers can see and discuss immediately.

2.  Engagement & Awareness Tool: When done collaboratively (e.g., peer-to-peer), observations can raise safety awareness, prompt conversations, and involve workers directly in safety discussions at the operational level.

3.  Identifies Specific At-Risk Behaviors: They can pinpoint recurring unsafe actions linked to specific tasks or locations, allowing for targeted coaching, refresher training, or procedural reviews.

4.  Cultural Signal: Implementing an observation program signals (internally and externally) that the organization prioritizes safety and is actively "monitoring" it, fulfilling a real or perceived regulatory requirement (note, some could argue this is more a criticism than a benefit).

5. Fits Existing Management Paradigms: It aligns with traditional top-down management approaches focused on individual performance and compliance, making it quicker, easier and cheaper to implement than complex systemic changes.

6. Gateway to Deeper Analysis: For some organizations, starting with behavioral observations is a first step towards safety engagement, potentially paving the way for more sophisticated analyses of systemic factors at a later date.

In my experience, behavioral observations can offer a relatively simple, actionable, and visible method to demonstrate safety activity, engage workers (albeit superficially), and address some immediate hazards, even if they don't solve deeper systemic issues. Their persistence often stems from practicality, and the behavioral tradition of focusing on observable actions. Nevertheless, I have witnessed several potential downsides and risks associated with behavioral observations that leaders would do well to consider, and seek to mitigate:

The risks of Behavioral Observations

1.  Fundamental Attribution Error & Victim-Blaming:    

Core Criticism: Overemphasizes individual actions while downplaying the crucial role of systemic factors (poor equipment design, unrealistic production pressures, inadequate maintenance, flawed procedures, insufficient training, poor management decisions).

Consequence: Can devolve into blaming the worker ("human error") rather than fixing underlying organizational or engineering failures. This is demoralizing and counterproductive.

2.  Superficiality & Reductionism:    

Core Criticism: Reduces complex safety performance to observable, often simplistic, "safe/unsafe" acts. Fails to capture cognitive processes (decision-making, risk perception, situational awareness), mental workload, fatigue, stress (psychosocial safety), and positive intentions behind actions.

Consequence: Interventions become superficial (e.g., "remind workers to follow procedure") rather than addressing why procedures are sometimes hard to follow or why “non-compliant” acts made sense when under pressure.

3.  Erosion of Trust & Psychological Safety:    

Core Criticism: Observations, especially by management/supervisors, can be perceived as surveillance or "catching people out." This undermines trust and discourages open reporting of near misses, concerns, or procedural difficulties for fear of blame.    

Consequence: Creates a culture of fear and silence, hindering proactive hazard identification and learning from incidents – the exact opposite of a mature safety culture.

4.  The Hawthorne Effect & Artificial Behavior:    

Core Criticism: Workers often modify their behavior because they know they are being observed (the Hawthorne Effect). This provides a distorted snapshot of work practices, not necessarily routine behavior.    

Consequence: Data collected may not reflect real-world risks or the effectiveness of interventions when observers aren't present, leading to false conclusions about safety levels.

5.  Subjectivity & Bias:    

Core Criticism: Identifying and categorizing behaviors is inherently subjective. Observers bring personal biases (confirmation bias, halo/horn effect) which can taint their interpretations of rules/situations.    

Consequence: Data reliability and validity are compromised. Observations may unfairly target specific individuals, groups, or shifts, or overlook contextually appropriate adaptations.

6.  Neglect of Latent Conditions & "Work-as-Done":    

Core Criticism: Focuses on "work-as-imagined" (procedures) vs. "work-as-done" (actual practice). Observations often miss the adaptive behaviors workers use to cope with real-world constraints, pressures, and imperfect systems – which can be both a source of risk and resilience.    

Consequence: Interventions based solely on non-conformance with imagined work may fail, or even increase risk by removing necessary adaptations without fixing the underlying system flaws that necessitated them.

7.  Resource Intensity & Diminishing Returns:    

Core Criticism: Effective observation programs require significant time for training observers, conducting observations, recording data, analyzing results, and providing feedback. Maintaining consistency and coverage is challenging.    

Consequence: Resources might be diverted from more impactful systemic improvements (e.g., engineering controls, better maintenance, process redesign). Focus can shift to meeting observation quotas and KPIs rather than meaningful safety dialogue.

8.  Quantification Trap & Misleading Metrics:    

Core Criticism: An over-reliance on counting "safe acts" vs. "unsafe acts" creates vanity metrics that may not correlate with actual safety outcomes (injury rates, process safety incidents). It can incentivize focusing on easily observable, low-risk behaviors.    

Consequence: Provides a false sense of security ("Our safe act percentage is 98%!") while potentially missing critical high-consequence risks that are harder to observe directly or occur infrequently.

9.  Potential for Gaming & Disengagement:    

Core Criticism: If tied to incentives or performance metrics, workers and observers may "game" the system (e.g., focusing observations on low-risk areas, avoiding difficult conversations, recording only positive observations).    

Consequence: Undermines the program's integrity and purpose, leading to cynicism and disengagement from the workforce. (Note, in my experience, this can also be true for management “Safety Walks”).

10. Equity Concerns:    

Core Criticism: Frontline workers are disproportionately the subjects of observation, while managerial decisions, design choices, and resource allocation (systemic factors) are rarely subjected to the same level of scrutiny via behavioral observation.    

Consequence: Can create a perception of unfairness and reinforce a top-down "workers are the problem" mentality, hindering collaborative safety efforts. While behavioral observations can provide useful data points and opportunities for coaching, an over-reliance on this approach risks neglecting the complex socio-technical nature of safety.

My key concerns center on the potential for behavioral approaches to blame individuals, ignore systemic causes, distort reality, erode trust, provide misleading data, and consume resources that could be better spent on higher-level safety controls and fostering a genuine learning culture. A balanced, more eclectic approach, integrating behavioral observations with robust analysis of systems, procedures, equipment, and organizational leadership and culture is essential.

About the Author

Clive Lloyd is an Australian psychologist who assists high-hazard organisations to improve their safety performance through the development of trust and psychological safety. He is a director and principal consultant of GYST, and developer of the acclaimed CareFactor Program.

Clive was recently named among the top 5 Global thought leaders and influencers on Health & Safety by Thinkers360. He is the author of the Amazon best-selling book "Next Generation Safety Leadership: From Compliance to Care".

For further information about the CareFactor approach please contact us at:

• Email: info@gyst.com.au
• Website: www.gyst.com.au