DEI needs more than Implicit Bias Training…

The currently most frequently used approach to attempting to achieve DEI goals is the utilization of Implicit Bias Training.  This approach focuses on making employees aware about their own individual biases, but doesn’t go far enough to truly impact the organization. 

There are several problems with implicit bias training alone.  For one, research shows that it's extraordinarily difficult to avoid triggering stereotypes. The activation of a stereotype is automatic, but how it plays out in a given situation can be controlled. Second, research shows that changing of one's behavior doesn't necessarily depend on changing implicit biases. The third problem is even bigger: most implicit association test (a main tool in Implicit Bias Training) is focused on the psychological basis of bias without looking at what bias at work really looks like, and with little or no guidance on how to stop it. 

Due to the complexity and relative stability of bias, it is bigger than trainings or mentoring programs can solve.  To truly have impact, the best first step is to tackle bias in key systems like hiring, assignments, performance evaluations, and promotions.   Improving the quality of business systems helps the business. 

To truly cement organizational commitment to DEI, we need to stop focusing nearly exclusively on helping people navigate systems that remain fundamentally unfair; we need to change the systems.  People analytics uses your own data to identify weaknesses and vulnerabilities in your systems and culture  to enact true improvement; something Implicit Bias Training simply can not do. 

Changing ‘hearts and minds’,  through implicit bias training,  is merely one element to improving DEI outcomes in an organization.  The other two important legs of the stool are using your organizational data to enable the revamping of systems to weed out bias and build a culture that supports DEI. 

If a company faces diversity challenges, typically it's because bias is constantly being transmitted, day-by-day, through its basic business systems: such as the hiring process, performance evaluations, and the way access is granted to valued opportunities.  

Just like for all business problems, we need data about what's going wrong, and metrics to measure your progress in fixing it. We need to tackle DEI using the same basic business tools that would be used for any other major organizational problem.  Organizational problems are often rooted in organizational systems; we need to change the systems.  Instead of focusing on people's subconscious thoughts, we need to look at how people actually behave and the things that govern that behavior. 

At Datafied PAS, we focus on two different kinds of metrics: outcome metrics and process metrics.   Outcome metrics measure demographic groups representation at different levels and parts of the company. Outcome metrics can tell you whether you have a problem, but not why or how to fix it.  

For that you need process metrics, which provide a barrier analysis: what's going wrong and where. Process metrics can pinpoint the separate processes necessary to achieve the company's diversity goals, providing each unit with tools such as a hiring metric, a promotion metric, and a turnover metric.  Process metrics identify precisely what problem or problems you need to solve. 

“Addressing bias in systems, policies and organizational culture requires changing structures.  If you had a serious problem with your supply chain, you wouldn’t try to fix it with a heartfelt memo…”