What Are The Alternatives to Behavior Based Safety

By Phil La Duke


Last week I posted yet another criticism of Behavior Based Safety (BBS) and it drew the following comment

“Good morning Phil I hope all is well. The argument for and against Behavior-Based Safety is as old as the first implemented methodologies, yet it still persists in many different beneficial and strange forms. Some refer to incentive schemes as BBS, others just a psychology-based approach as BBS and others watch a video, read an article and attempt to make it work with widely ranging results on culture and performance. I believe BBS to be a situationally-appropriate tool for a small aspect of safety. Moreover, it should be a tool focused on better understanding performance and the influences on it, than an awareness or accountability mechanism. The latter tends to cause some of the problems you write about and I have seen as well. Rather than perpetuating the continuous critique, I would sincerely be interested in reviewing the specifics of the methodology/approach/tool you propose that will accomplish the same results in the small aspect of safety BBS benefits. Would you please share?”

This isn’t the first time I’ve been asked to share the alternatives to BBS. But this is no easy feat—first of all, as this comment suggests, there is far from a single source of truth that defines BBS and its elements. Before we can discuss the relative effectiveness of BBS we need to agree as to exactly what constitutes “effectiveness” of a safety management system.  The criteria I will use are:

  • Cost
  • Effort
  • Sustainability
  • Value

Most of the purveyors of BBS agree that the following are elements of a comprehensive Behavioral Safety management system:

  • Evaluation of Worker Behavior Using Checklists.

Trying to find a competitive system involves some modification of the behavioral observation.  Personally, I reject the idea that people get hurt because they knowingly and consciously behave in ways that put them in jeopardy.  I am supported in this belief by Joseph T. Hallinan, author of the book Why We Make Mistakes: How We Look Without Seeing, Forget Things In Seconds, and Are All Pretty Sure We Are Well Above Average; David Marx author of Whack A Mole—The Price We Pay For Expecting Perfection; and Zachary Shore the author of Blunder: Why Smart People Make Bad Decisions. Putting this philosophic difference aside, conducting periodic reviews of the work area that focuses on all the hazards instead of focusing on purely (or even chiefly on) behaviors is far more likely to lower the risk of process failures which not only endanger workers but also puts quality, through-put, and production at risk. What alternative is there to BBS? Several tools come to mind:

  1. Layered Process Audits. Layered Process Audits are checks conducted by various levels of management.  The primary purpose is to ensure that the process as performed conforms to the process requirements. Part of the Layered Process Audit system is the verification that all mistake proofing protections are in place and operational.  This is essentially an improved version of the behavioral observations that requires less effort, is far less costly, is relatively easy to sustain, and ultimately returns far more value than the behavioral observation.
  2. Kaizen Events. Kaizen events are ad hoc activities designed to improve the efficiency of the work area. Kaizen events involve the workers in the area who participate in improvements by identifying and eliminating sources of waste—including those things that are likely to cause injuries.
  3. 3.    5S Audits. 5S is a powerful tool designed to reduce process variation and make the work area more efficient and safer. It involves simple workplace reorganization that sorts, sets in order, scours, standardizes, and sustains improvements in the workplace.

Data Collection From Observations. Behavior Based Safety systems rely on measurements taken by watching workers perform their jobs.  While safety information should be routinely analyzed and trends should be studied to determine proactive initiatives, here again BBS falls short.  First, in an attempt to overcome Heisenberg’s Uncertainty Principle (which states the observation itself alters the factors being observed) BBS depends on many observationsTo achieve the desired number of observations BBS system relies on numerous  trained observers who observe their peers and provide them feedback on their behaviors.  Such activities rarely normalize the data to adjust for the Hawthorn Effect (the tendency for workers to improve simply because the organization is taking action).  Data collection relative to safety indicators are key to constructing a coherent safety strategy, but again, there are better (as it relates to Cost, Effort, Sustainability and Value). Let’s take each of these factors one at a time.  The cost of conducting observations are substantial—checklists must be constructed, workers must be trained both in evaluating and being evaluated, and the observers must be paid a wage to conduct the assessment—the effort (for the same reasons) is onerous; and the long-term sustainability of these activities is dubious.  Add to this the resentment and morale issues (up to and including Union drives or labor unrest) associated with peer-to-peer audits and you have a really bureaucratic system that creates head count and saps productivity. A far better solution is the balanced score card.   According to http://www.balancedscorecard.org/BSCResources/AbouttheBalancedScorecard/tabid/55/Default.aspx the balanced scorecard is “a strategic planning and management system … It was originated by Drs. Robert Kaplan (Harvard Business School) and David Norton as a performance measurement framework that gave managers and executives a more ‘balanced’ view of organizational performance.”  General Electric, hailed by many as a pioneer of modern management, was an early adopter of the balanced scorecard approach

Modern balanced scorecards has evolved beyond being a simple performance measurement system and has blossomed into to a strategic tool for planning and managing a business; it provides a visual representation of the progress against strategic initiatives.  Typically the balanced scorecard provides a framework for achieving goals in Safety, Quality, Delivery, Morale, and Environment (SQDCM). In addition to the obvious advantage of leveraging existing efforts and collaborating across multiple functions, the balanced scorecard imbeds safety into standard operating procedures—instead of acting as if safety was a discrete element it is rightfully treated as a pillar on which an effective organization is built.

  • Worker participation. An oft-cited reason for the “success” of BBS that it fully engages the workforce in safety management. Proponents of BBS extoll the virtues of this grass roots approach as opposed to a management driven top-down approach.  Many systems include incentives for making the workplace safer—safety BINGOs, bonuses for injury-free quarters, or similar initiatives.  A more economical, holistic approach is to use existing employee suggestion programs to solicit and reward ideas that genuinely improve the safety of the workplace.  It may not be sexy, but why create a parallel process that is limited solely to worker safety when a larger, more inclusive system already exists?
  • Focus On Specific Unsafe Behaviors.  BBS proponents tout the relatively scarce types of behaviors that cause the majority of injuries.  Here I believe that this is not in fact, BBS.  This is an attempt to use process based safety tools to address a shortcoming of BBS.  But let’s take a quick look at the practice of using Pareto chart analyses to target the behaviors of greatest risk. Pareto charts track quantitative (counted) data and not qualitative (measured) data.  This kind of data is generally (but not necessarily) derived from Area Maps or Body Maps.  Since the severity of the event is not collected in Pareto analysis (the data is assumed to be more or less the same severity and holds more or less the same risk of injury) it is inappropriate to use this data in determining the critical few behaviors that represent the greatest danger.  Furthermore, this type of analysis essentially ignores hazards that are largely environmental, organizational, or mechanical.  Instead of this approach organizations should focus on ALL hazards instead of focusing on behaviors.
  • Focused Feedback Performance.  In BBS feedback usually takes one of three forms: feedback at the time of observation; graphs of trends used in weekly discussion with work crews, and monthly discussions about safety by management.   While feedback on behavior is valuable it only can provide benefit in cases where the behavior was deliberate.  A large percentage of unsafe behavior is simple human error and no amount of feedback will change the fact that people make mistakes.  Another substantial source of unsafe behavior is behavioral drift (the tendency to slowly move away from the standard procedure until one unknowingly moves into risky behavior).  According to David Marx (in his book Whack A Mole: The Price We Pay For Expecting Perfection) contends that drift is an unavoidable part of human behavior.  Here again, telling someone that they drifted isn’t all that useful in changing worker behavior.  Finally, there is reckless behavior (defined as the willing choice to behave in a way so risky that no reasonable person could ever defend the behavior as in proportion to the perceived reward.) In cases of recklessness feedback on the behavior is unlikely to result in behavioral improvement.  Instead of focused feedback on behavior organizations would be far better served by implementing a Just Culture approach to safety combined with training in decision-making and a comprehensive program of error proofing.
  • Data-driven decision-making. How is this a differentiator of BBS? Aren’t all safety management system based, at least to some degree, based on data-driven decision making? Clearly making the decisions on data that is primarily based on behavioral data at the expense of other relative factors is not something to brag about.
  • Requires visible on-going support from managers and front-line supervision. Here again, this is not something limited to BBS. (And a contradiction with BBS’s claim that its success is rooted in the commitment of the workers and a grass-roots movement.) If managers and front-line supervision aren’t supportive of safety it will most certainly fail irrespective if it is a BBS, process based, continuous improvement, or Just Culture approach.. But in all cases I think this is a cop-out.  A good safety system should begin by engaging leadership before it starts laying out commandments.  When I custom-design a safety system I begin by assessing the leadership’s commitment to making the changes necessary to effect lasting change.  If I am not satisfied that this commitment exists I walk away. In my one-year engagements I have completely transformed cultures and produced for my customers sustainable and effective safety systems that they own without creating a parasitic relationship between vendor and customer.

This is just the tip of the iceberg—I didn’t even touch the many companies who make money by certifying people to perpetuate their crappy safety systems, sell “training materials” or dozens of trademarked consumable add-ons that end up unnecessarily costing the customer tens of thousands of dollars annually.

Did you like this post? Did you find it helpful? Was it thought provoking? Why not share it with your peers? I think they would appreciate it and I certainly would.