Safety Leadership: Three truths and a lie about analytics and safety

Editor’s Note: Achieving and sustaining an injury-free workplace demands strong leadership. Throughout 2013, experts from Ojai, CA-based consulting firm BST will share their point of view on what leaders need to know to guide their organizations to achieve world-class safety performance.

Big data. Data mining. Predictive analytics. The buzz surrounding analytics has grown almost as much as the data itself. Over the past 10 years, there has been a tremendous increase both in the sheer amount of data collected (by some estimates growing 30 to 40 percent per year) and in the number of tools available to analyze and draw insight from this information. This raises the question: Is analytics the game-changer it’s made out to be?

Our experience is that the field of analytics, which is simply statistics applied to the improvement of business insight and decision-making, offers promise for safety applications. But as with any new technology, leaders must learn to distinguish help from hype. In our work, we’ve discerned three truths (and one misconception) that can serve as a guide in assessing safety analytics for your organization.

Truth 1. Today’s technology presents new opportunities

Although there is nothing new about using data to understand our world and predict future events, new technology has exponentially increased the amount of data – and the ways we can use it. Just as some companies use analytical tools to track and anticipate consumer behaviors and preferences, organizations can now use analytical tools to track and anticipate injuries, monitor exposure patterns, and deploy resources and training more strategically. Analytics is not a magic bullet, however. Leveraging these advancements requires that leaders throughout the organization know the safety principles and concepts underneath the data.

Truth 2. Presentation matters

In many organizations, critical safety data is buried deep within spreadsheets or delivered in a way that assumes leaders already know the context (or the urgency). With the growth of sophisticated data presentation tools, analysts can now create heat maps of exposures across a site or build instant models comparing key variables – effectively changing how people interact with information. The challenge for analysts and safety professionals is to learn what configurations and tools will help various stakeholders “see” connections and make effective decisions.

Truth 3. Data quantity and quality is critical

Statistical models are only as good as the data that goes into them. However, analysts have a tendency to focus on creating the best models with whatever data is available, even if that data is flawed or lacking in key variables. This is analogous to the old joke about the man looking for his lost keys at night. He doesn’t look for them in the shadows where he dropped them but under a light pole because that’s where he can see better. Even though the analysis may look impressive, leaders must ask questions about the data that goes into the model and, more important, what information is missing because it was too difficult to collect.

The lie: Prediction is power

It can be easy to get caught up in the excitement of new technology. But it’s not the ability to predict relationships that creates value – it’s the ability to drive effective action. A finding that shows 80 percent of incidents involve men even though the company is only 60 percent male is interesting but not terribly actionable. On the other hand, a model that shows that a significant shift in safe observations is a reliable predictor of incidents is helpful because it enables you to respond with concrete action.

Power and potential

The promise of Big Data for safety is real. But to work, analytics must be rooted in sound safety and management disciplines. Assessing data collection tools and processes, and redefining the information needed, can be challenging. But with work, these tools can reveal relationships previously not understood and enable an organization to take action that helps drive down incidents.

This article represents the views of the author and should not be construed as a National Safety Council endorsement.

Michael Mangan is director of research and development for BST. Mangan’s work helps organizations understand performance and design interventions for improvement.