The Campbell Institute

The Campbell Institute: Safety leading indicators: A big data problem (part one)

The Campbell Institute

The Campbell Institute at the National Safety Council is the EHS center of excellence. Built on the belief that EHS is at the core of business vitality, the Institute seeks to help organizations, of all sizes and sectors, achieve and sustain excellence. Learn more at

Companies that treat safety leading indicators as a “Big Data” problem are applying analytics techniques to steward the indicators that truly do lead to better safety outcome performance. They’ve decided to end a long stretch of time during which senior leaders have influenced behaviors by reacting strictly to lagging performance outcomes.

Given the explosion in data quantity over the past decade, the term “Big Data” has grown in prevalence. It’s an ambiguous term. Some refer to the term when alluding to an inordinately large set of structured relational data. Others consider the term to strictly mean data sets so large and complex that traditional data processing technologies are inadequate. In almost all cases, the mention of “Big Data” infers the use of analytics techniques to glean valuable insights from the data.

It’s this context of applying analytics to safety data that is the focus of this column. Companies that use this approach believe shifting the attention of senior leaders to focus upstream of lagging outcomes is a big problem and one that can be addressed with data – i.e., a big “data problem.”

The current state of safety data in most companies entails some site-level use of non-lagging measurements (often called leading indicators), but these metrics are typically NOT vetted with data and analytics for proving their merit to “lead” to better performance and are usually NOT rolled up to a regional, divisional or enterprise-level for senior leader stewardship.

The issue of calling workforce attention to non-vetted leading indicators is akin to a doctor prescribing drugs that have not been proven to work. Ensuring this doesn’t happen is a key reason for the Food and Drug Administration. But we don’t have an FDA for safety leading indicators. So companies are on their own to prove the effectiveness of a leading indicator before they prescribe it as medicine for safety performance.

The issue of leading indicator “roll-up” sounds administrative in nature but it’s bigger than that. Top business leaders love to use numbers to drive their message. In lieu of leading indicators that roll-up, these leaders use the only numbers available – lagging metrics. Many papers have been written to attest to the perils of driving through a rear view only – so there’s no need to expand here.

It’s not a new idea to apply statistical methods for identifying the correlation between leading variables in safety and lagging safety performance outcomes. This has been done for years to convey the value of workforce perception surveys, behavioral observation programs, management system assessments, etc. So, what has changed?

For decades, companies have deployed many of the same activities in effort to improve safety performance (e.g., near-miss reporting, incident investigations, corrective/preventive actions, behavioral observations, management of change, risk assessment, HSE management systems, and on and on). Historically the data from these leading activities was “practically” inaccessible at an aggregate level. It was either manually collected or tucked away into disparate site-level systems.

However, over the past couple of decades there has been a movement toward enterprisewide information systems to enable both the local execution of these activities and the aggregate collection of this leading data. Now, large individual companies and collections of smaller companies are efficiently producing data sets big enough for applying analytics techniques to identify the measurable characteristics of executing safety activities that lead to better safety outcome performance – real leading indicators.

Given the information system platform as the source of data for these indicators, these companies are able to leverage the same platform to deliver Big Data-vetted measurements for top management to steward the changes relevant to improving performance throughout their organization. They are on the path to solving the “Safety Leading Indicator Big Data Problem.”

(Part two of this piece will appear in the August issue of Safety+Health.)

Sign up to be notified by email about new Campbell Institute columns.

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

Joe Stough leads the Operational Excellence Advisory Services group at IHS, a member of the Campbell Institute. IHS uses structured analytics and research methodology to reveal the measurable factors that drive operational excellence.

Post a comment to this article

Safety+Health welcomes comments that promote respectful dialogue. Please stay on topic. Comments that contain personal attacks, profanity or abusive language – or those aggressively promoting products or services – will be removed. We reserve the right to determine which comments violate our comment policy. (Anonymous comments are welcome; merely skip the “name” field in the comment box. An email address is required but will not be included with your comment.)