The 1st article in this series discussed descriptive analytics and how it describes “What happened?”  In this post, I will address diagnostic analytics which answers the question: “Why did it happen?” Now you can begin to explore the relationship between variables/trends to try to understand why and how performance indicators move together (are positively correlated) or move in opposite directions (negatively correlated.)

As previously mentioned, each successive level of analytics is more sophisticated and higher on the value chain, but also introduces some uncertainty the more complex and the less easily predicted.  In diagnostic analytics, we are trying to understand how trends in variables impact one another.  For example, in descriptive analytics, we were able to understand general characteristics of the employee population such as average age and distribution of age by generation (i.e., Baby Boomer, GenX, Millennial, GenZ).  We can also learn something about the average tenure of the employees.  In diagnostic analytics, we are trying to learn something more complex about the workforce .

In this specific example, we may want to learn something about the average tenure by generation.  We might discover when we dig deeper into the data that the longest average tenure is for the Baby Boomers, and the lowest tenure is for the Millennials.  Once we learn this, we have actionable information that we didn’t have before and if tenure is important to the organization, then we know that we need to focus our attention on retaining the Millennials.  We might even want to explore the trend among the Millennials to see if there is a certain division that is doing better than others in retaining the Millennials. Can we learn something about the management style of that division to understand how to adopt it to other divisions where average tenure is lower?

This is only one example, but you can apply this concept of looking at the movement of multiple trends concurrently to understand if impact each other. 

Some other relationships that may be interesting include:

  • Employee productivity and profitability
  • Employee productivity and attrition
  • Expense factor and profitability
  • Mobility and attrition rate
  • Recruitment efficiency and first year attrition rate 

As you can see, key performance indicators, expressed as ratios and tracked over time, can provide solid information that can inform your decision making.

Please join me for the third post in this series which will deal with Diagnostic Analytics as we continue along the path to derive a relationship between people data and bottom-line returns. Now that we know what happened, can we use data to ascertain why it happened?

Solange Charas is a senior-level human resources expert with 30+ years of experience as a consultant, practice leader, top corporate executive, and board director across all industry sectors.   She was the Chief Human Resources officer at Havas Worldwide, Benfield and Praetorian Financial Services Group and held senior-level positions at Ernst & Young and Arthur Andersen.  She serves of the boards of 2 public companies, a non-profit organization and a higher-education institution.  She is the Founder and CEO of HCMoneyball – a SaaS company founded to provide support for enhanced decision making about spend on people in any organization.

Solange earned a PhD in Management from Case Western Reserve University’s Weatherhead School of Management, an MBA in Accounting and Finance from Cornell University’s Johnson Graduate School of Management, and a BA in International Economics from the University of California, Berkeley. She has authored numerous articles, including “The Art and Science of Valuing People” in HR Director, “6 Ways to Coach Your Company’s Teams to Be Champions” in Entrepreneur Magazine and “Why Men Have More Help Getting to the C-Suite” in Harvard Business Review.