Reducing Voluntary Turnover Through Predictive Analytics | Deloitte US has been added to your bookmarks.
Reducing voluntary turnover through predictive analytics
Finding the balance in workforce retention
A large, global pharmaceutical company was having trouble with its China market sales force. The problem was two-fold: the company had issues retaining high-performing salespeople, and—despite having problems with retention—the company needed to double its sales force to meet customer demand and increase market share. They had no specific retention strategy, and they had little insight with which to develop one. The Deloitte workforce analytics team assisted them in developing an analytics-based predictive retention strategy and solution.
The challenge was steep, but not insurmountable. For the first phase of the project, management chose to focus on developing a retention strategy. They wanted more insight into who their highest-performing sales employees were across China, and they wanted to understand what motivated those employees to stay or go. They also wanted to build a retention strategy that would outline concrete actions and talent solutions to help reduce turnover. However, despite the need to reduce turnover and hire new employees, the company had to carefully allocate limited financial resources to solve the problem. In essence, they needed to do a lot with a little.
Developing roadmap strategy for workforce retention
Deloitte assisted the company in using analytics to build a predictive application to provide management with the insight to implement targeted retention initiatives for the sales force. The team used statistical modeling and visualization software to develop a retention solution that would drive innovation and insight through the power of data.
Using three years of employee data, the team built a model to cross-reference data to predict attrition risk by individual employee. They conducted initial meetings with the client to develop a variety of diagnostic factors that would aid in predicting those employees who were likely to leave. The factors included obvious things like present and past work performance, span of control, level, benefits, and salary. Other, not so obvious factors, such as supervisor performance, were also included.
The model helped the company assign a retention score—from 1 to 100—to each sales employee in the sales function. Thus, management could effectively target both high-performing and high-risk employees and develop a retention strategy to keep those valuable resources from walking out the door.
The solution included visualization capabilities to analyze projected turnover by region, employee, and a series of other dimensions. For instance, a manager could hover over the Shanghai location and drill down through a list of employees—and their risk scores—to determine which employees to target for retention efforts.
Impact on the business
The solution had an enormous positive impact on the business. With the insights gleaned through the application, management was able to develop an effective employee retention strategy. The strategy facilitated more efficient allocation of scarce resources through the identification of those retention initiatives that had the highest value and impact.
As a result, voluntary and total turnover, and loss of productivity for pivotal roles were reduced significantly. In fact, one year later, the company had a 98% retention rate in that critical employee group. As an added benefit going forward, the company was also able to use its enhanced analytics capabilities to boost its pipeline of new talent—thus getting a leg up on the next area of focus: doubling its sales force.