Predictive Analytics Times
Predictive Analytics Times
Machine Learning and Artificial Intelligence: Not New Concepts for the Data Science Practitioner
 Economic disruption is a reality which has been a...
Interview: The Institute of Business Forecasting & Planning Talks to Dr. Eric Siegel
  Dr. Eric Siegel cuts through the buzzwords surrounding...
Investment Modeling Grounded In Data Science
 For more from Dr. Elder, join Predictive Analytics World...
Some Thoughts On Being a Data Science Entrepreneur in a Disruptive Economy
 The movie “Being There” may seem like an odd...

6 months ago
Wise Practitioner – Predictive Analytics Interview Series: Haig Nalbantian & Tauseef Rahman at Mercer


By: Eric Siegel, Founder, Predictive Analytics World for Business

In anticipation of their upcoming conference presentation at Predictive Analytics World for Business Las Vegas, June 16-20, 2019, we asked Haig Nalbantian, Senior Partner, Co-leader Mercer Workforce Sciences Institute & Tauseef Rahman, Principal, Workforce Strategy & Analytics at Mercer, a few questions about their deployment of predictive analytics. Catch a glimpse of their presentation, How to Link Compensation to Organization Design to Optimize Career Rewards and Workforce Cost: The Role of Advanced Analytics, and see what’s in store at the PAW Business conference in Las Vegas.

Q: In your work with predictive analytics, what behavior or outcome do your models predict?

Haig Nalbantian: With respect to “career rewards,” the topic of our upcoming PAW Business conference presentation, there are two key outcomes of concern that advanced analytics can help organizations better understand: (1) voluntary turnover and (2) “career value.”

Working with client organizations, we routinely model the drivers or “predictors” of voluntary employee turnover. Such models identify and quantify the factors that most influence the employee’s decision to stay or leave the organization. As this decision amounts to “voting with your feet,” the models can help uncover what employees actually value in the employment relationship, something of value to employers trying to optimize their Total Rewards package or broader Employee Value Proposition (EVP). Among the most important learnings that arise from this work concerns the relative value of the career components of rewards as compared to the “here and now” of pay and benefits as well as those dimensions of career rewards – advancement, lateral moves, learning, development, supervisor and team relationships –  that have the greatest impact.

The second outcome on which we focus with respect to career rewards is a quantitative indicator of what we call, “career value,” defined as the present value of an employee staying and advancing in their organization over time. Specifically, we draw on employee data to calculate the “returns to promotion” from one career level to the next in a way that reflects promotion probabilities at each career level, associated pay differentials and turnover probabilities as well. The resulting measure identifies the strength of career rewards and potential misalignments in the trajectory of career value across the career hierarchy. When joined with turnover modeling, quantification of career value can help employers gauge the overall strength of the career component of rewards and the extent to which it is helping them secure the right workforce and motivate high performance. Those two objectives are at the heart of an organization’s people strategy. And achieving the right balance between the “here and now” of pay and benefits with the “career rewards” parts of the employment deal is critical to an effective people strategy

Q: How does predictive analytics deliver value at your organization – what is one specific way in which it actively drives decisions or operations?

Haig Nalbantian: Too often, decisions about rewards are based on market surveys of rewards “competitiveness” and other benchmark comparisons of reward best practices. These are important inputs, but they can fall short for two reasons: (1) they usually fail to capture and measure certain key dimensions of career rewards that help define the value and competitiveness of employment in different organizations, for example, the speed and value of promotion, the nature of work assignments and/or whom employees report to or work with, learning and development opportunities, among other things. (2) They don’t speak to the actual impact of reward components on employee behavior and/or business performance. Predictive analytics can help address these gaps.

So, for example, in a well-known global pharmaceutical firm, predictive analytics revealed that remarkably strong and well-aligned career rewards, including a generous Defined Benefit (DB) Pension Plan, were functioning effectively to support execution of a “build-from-within” talent strategy. Rewards were tightly linked to advancement up the hierarchy with almost perfect calibration between promotion rates and associated pay changes. The DB plan worked to provide inducements for employees to stay and grow with the firm during the middle stages of a career as well as incentives to retire “on time,” an essential component of a reward system that values tenure by back-loading pay and benefits. Not surprisingly, turnover modeling revealed that the career components of reward completely dominated pay elements – e.g., base pay level, bonus payouts, LTI – as predictors of retention. Employees were clearly responsive to career value indicating that the firm’s reward strategy was working hand-in-glove with their talent strategy to help deliver and motivate the right workforce.

With such strong evidence in hand, the company’s HR and Benefits leaders were able to convince their Board of Directors that following the herd in the rush to abandon their DB pension plan, without anticipating and dealing with the likely unintended negative consequences, would be self-defeating from a financial and risk perspective. Action on the DB plan was tabled.

Q: Can you describe a quantitative result, such as the predictive lift of your model or the ROI of an analytics initiative?

Haig Nalbantian: Strong career rewards are important to the extent that an organization derives value from length of service and the associated institutional knowledge. How can we quantity that value and be sure that such “firm-specific human capital” is actually a driver of business value? Here is where the use of predictive analytics can be especially powerful. Applied to longitudinal workforce and business performance data, our Business Impact Modeling® methodology can be used to statistically estimate the relationship between employee tenure and various measures of business performance. So, for instance, in a large professional services firm we found that average tenure among sales and delivery teams was the single biggest driver of year-to-year revenue growth from their large national clients; specifically, an additional two years of average tenure corresponded to more than 15% higher revenue growth from their clients, all else being equal.

With this kind of evidence in hand, the leadership of this company had little difficulty appreciating the importance of ensuring a strong and aligned structure of career rewards to help build a tenured professional workforce.

Q: What surprising discovery or insight have you unearthed in your data?

Tauseef Rahman: Fundamental to the career rewards concept are the career movements and relationships themselves, such as promotions, or the span of control of one’s own manager. In this regard, organization design plays a fundamental part in explaining what career rewards are possible. The advances in our data and research allowed us to quantify the shape of an organization in a standardized fashion.

For example, some organizations are pyramid shaped and have a strong hierarchy, thereby conveying lower and lower probabilities of promotion as you move up the organization in order to maintain the pyramid. Other organizations may be more block-shaped, with the workforce equally distributed at all levels of the organization, implying that the likelihood of moving up from one level to the next doesn’t decrease.

The core insight is that organization design itself should be calibrated with pay to optimize the impact of career rewards. If the career reward model is like a tournament, where promotion is leveraged as relatively few people advance, a pyramid shaped organization would better facilitate that outcome. Therefore, standardizing the quantification of organization shape is useful in helping us understand how organization designs vary from industry to industry, and even within industry, thereby enabling better translation of analytics to meaningful program changes.

Q: Sneak preview: Please tell us a take-away that you will provide during your talk at Predictive Analytics World.

Tauseef Rahman: We will share some of the findings of our career rewards analytics that show the relative strength and impact of the “here and now” of pay and benefits compared to “career rewards”, and how different organization designs influence and are influenced by career rewards. More importantly, we’ll share our approach for quantifying organization shape. This “shape score” can prove extremely valuable in putting into context the people analytics organizations run.



Leave a Reply

Pin It on Pinterest

Share This