AMSJ » The Carrot vs. the Stick – Feedback in Behaviour Based Safety Programs
LATEST NEWS Psychosocial Risks Safety Consultants Safety Reporting Systems

The Carrot vs. the Stick – Feedback in Behaviour Based Safety Programs

behavioural safety

Behaviour based safety is an advanced safety management technique employing observation/feedback sessions as the primary motivator for behaviour change. Behaviour in safety management refers to the choices people make to work safely or take risks. These choices are made from a complex series of decisions, many of which are unconscious or based on employee training or mentoring. The primary motivator in successful behavioural safety programs needs to be positive reinforcement and mutual respect. However, experience shows that negative reinforcement may be more perception than reality in properly designed behaviour based safety systems. In this article we examine the positive and negative aspects of behavioural based safety and how organisations can use it to improve safety performance.

Behaviour-based safety (BBS) programs have evolved from early research in the 1930s by Herbert William Heinrich of the Traveler’s Insurance Company. Heinrich studied extensive accident records and concluded that a large percentage of accidents are caused by human error. During this time, B.F. Skinner was also doing pioneering research on behavior in animals and humans. Skinner’s work was pioneering because he applied a scientific method to behavioral stimulus and response. The BBS approach is arguably founded on behavioral science as conceptualised by B. F. Skinner (1938, 1953, 1974). The work of these two researchers has gradually evolved into a safety management system called behavior-based safety.

The term “behavior-based safety” is credited to consultant psychologist Dr. E. Scott Geller, who first used the term in 1979.


Behavior-based safety is sold by many consulting companies in a variety of forms. The basic structure of a behavior-based system is consistent with all systems.

Behavioural safety choices
BBS safety can be approached from negative and positive aspect.
  • An observation is performed on an employee performing a work task. Ideally, the observation is a pre-arranged appointment, not a surprise visit.
  • The observer looks for both safe and unsafe behaviors. Safe behaviors are behaviors that minimize risk to the employee. Unsafe behaviors are behaviors or actions that have a component that either violates safety procedures or incurs risk, such as not using personal protective equipment or inspecting the equipment prior to use.
  • The observer records the observation and provides a feedback session to the employee. The focus of the feedback session is to acknowledge the safe behaviors already being used by the employee. This is the “carrot”. The observer also notes one or two unsafe behaviors, coaches the employee on why these behaviors are unsafe, and asks the employee for agreement to “unlearn” the behavior. 
  • The observer records the observation and discussion, with emphasis on the positive behaviors seen and any unsafe behaviors with a path forward. Any safety concerns from the employee, such as incorrect equipment procedures or maintenance problems, must be documented.
  • The employee and observer may agree to meet again. The observer should commit to tell the employee if any of the employee’s safety concerns are corrected.

If this is the essence of a behavior-based safety program, where is the “stick”? The best answer is – that depends. There are a few potential pitfalls to an observation-based safety program.

Problems of behaviour based safety programs

Some employees think of behaviour-based safety observations as an audit, a term historically connected in safety to an observer that is determined to find something wrong. Bad auditing has plagued industry for many years, resulting in employee discipline and punitive actions. This negative baggage, in some circumstances, now hangs over all safety programs as soon as “an observation process” is mentioned.

In the past we have observed many companies having initiated observation-based safety programs without proper observer training by professional consultants. Positive reinforcement-based observation is not an innate skill and requires coaching to remove the negative reinforcement that many employees and supervisors consider a normal part of safety observations. All employees in an organization must complete observer training before the observation phase begins.

One of the biggest pitfalls may be a company’s management. Management may be resistant to the idea of employees taking time to observe each other and “pat themselves on the back” for being safe. A successful behavior-based safety program must have buy-in from all levels of an organization, including management and executive staff. All employees must participate, and no one can be exempt from an observation. How do you observe the company CEO in a safe office environment? Take them out of the office; invite them to do a safety walkabout and see how they manage safety regulations and interactions with other employees.

Some unions throughout Australia have voiced strong disapproval of behavior-based safety systems. A common complaint is that a program to blindly follow safety regulations is manipulating the employees. Published articles also claim the system is flawed because an observed employee will only work safely while they are observed. The reasons for union disapproval are complex; some of the concern may stem from the negative aspects of Skinner’s stimulus/response research. 

Unions concerns with Behaviour based safety

In 2005, the MUA defined13 aspects of Behaviour based safety programs and the basis for their disapproval. These included

  1. Observing others’ behaviours focuses on the end of the chain of events that lead to a worker’s actions…………..just look at any proper accident or incident investigation root cause. BBS promotes the lie that accidents are caused by one event, when all the research shows a chain of events leads to accidents,
  2. Observing workers’ behaviours does not ask the questions “why” is s/he doing that job that way?; why is s/he doing that job?; why is s/he using that chemical? etc,
  3. Observing others only looks at what happens often and repeatedly e.g. breakdowns, jams or quick production runs or days when staff are short are not taken into account. These programs miss any unusual or complex unsafe events.
  4. Observing what workers are doing will not give any information about the inherent dangers in a work process. Observing someone’s behaviour does not give any information about the effects of:
    • fumes e.g.benzene which causes cancer,
    • dusts e.g. asbestos , ceramic fibres, silica, cadmuim,
    • epoxy resins or two part paints or hair dyes causing asthma
    • biological hazards, infections,
  1. Observing workers and changing their behaviour can only change the decision making at the lower level of decision making. It is very hard to affect the big decisions without asking “why”. See 2. above.
  2. Observing workers takes lots of time and costs a lot. Over time this often crowds out other health and safety activity e.g. focussing on hazards at source and implementation of hierarchy of control approaches,
  3. As one H&S expert1 said– rather than these programs being the cherry on top of a multifaceted OHS program, BBS can crowd other approaches out and end up being the poison garnish that ruins the whole dish,
  4. Observing what workers are doing, does not observe what managers or the Board are doing or not doing. Managers, owners or the Board are the ones with the power and access to resources to make the necessary changes for improved conditions. This is recognised as the duty of care in both common law and OH&S law,
  5. Observing workers working with clients cannot answer questions about factors which lead to stress e.g. fatigue, emotional pressures, anxiety, workload, decision making violence and aggression, bullying,

10.Observing what workers are doing, tends to blame the worker, even if the program tries not to e.g. even if it is anonymous, voluntary etc. It perpetuates the myth of the careless worker,

11.Observing workers can put worker against worker,

12.Observing workers does not try to answer the question – if we moved “Bill” from this job would someone else be likely to be injured, because when we focus on Bill we often cannot see beyond him. If it is possible that someone else could be injured then the problem is NOT with Bill’s individual behaviour, but with the job or the system, or something else,

13.Observing workers does not question those who make decisions about how production is designed, the workplace is organised or laid out etc. Managers, owners or the Board make those decisions.

Factors influencing human behaviours
Factors influencing human behaviours

A stick approach can be damaging

These issues notwithstanding, there should not be a “stick” in a behaviour-based safety program using positive reinforcement.

Only two issues should stop an observation/feedback session; a behavior that poses an immediate risk of injury or death to the employee or others, and a wanton or deliberate disregard for established safety procedures. The issue of a risk behavior can still result in a positive reinforcement session; the employee should be advised of the behavior, and the employee and observer can work together to create a safe behaviour.

If a risk behavior results from improper employee training, management should be notified immediately so corrections can be make to procedures and training. Deliberate disregard for safety procedures, particularly if admitted by the employee, should be referred to management.  The observer should not lecture or discipline the employee; their only action should be to stop the work task until further investigation is completed.

Good safety programs must include a variety of topics, activities and recognitions to keep employees engaged. In a company culture where employees can accept observation and feedback, a behaviour-based safety program can drive incidents and injuries to extremely low levels, and raise employee ownership of safety if it is implemented effectively…and that is the key issue in any safety program.

If you are about to embark on a behaviour based safety program, you should talk to different consultants familiar with behavior-based safety programs, and look for a system that best suits your company and culture.

Safety meetings and incident reporting are necessary functions of any safety program, but look for ways to provide positive feedback for safe employee behaviors.


Geller ES. (2005) Behavior-based safety and occupational risk management. Behav Modif. 2005 May;29(3):539-61. [Available online at], Date accessed 11/1/2020.

Skinner, B. F. (1938). The behavior of organisms: An experimental analysis. Acton, MA: Copley.

Skinner, B. F. (1938). The behavior of organisms: An experimental analysis. Acton, MA: Copley.

Skinner, B. F. (1974). About behaviorism. New York: Alfred A. Knopf.

Add Comment

Click here to post a comment