Using Predictive Analytics to Reduce Employee Turnover

Banner for Predictive Analytics Post

This article is an abstract of an ADP Research Institute report entitled Revelations from Workforce Turnover.

Analyzing Employee Turnover

High turnover is a business liability. That’s intuitive, but according to new research by ADP Research Institute, a generalized approach to solving employee turnover problems isn’t particularly effective. In a new report, ADP employs data analytics to dig into the factors that produce voluntary turnover and to create a modeling system that can be predictive for individual enterprises.

According to the report, traditional attempts to analyze turnover don’t produce specific and actionable insights:

Merely suggesting increased dissatisfaction when new jobs are readily available, or evoking  the value of orientation and engagement is one thing. Being able to map an array of influences and then point to the combinations driving turnover is quite another.1

The ADP study evaluated a large set of payroll data from 41,000 companies and 12.5 million employees to assess the factors that contribute to voluntary turnover and to create a predictive model that can be used by companies to improve their performance.  The data was used to calculate seasonality and patterns of turnover by industry.

ADP RI Graph

Surprisingly, the majority of firms across all of the industry segments fall below industry averages for turnover. The report suggests that predictive modeling can be used to refine and improve retention strategies in successful organizations as well as to identify the problems in companies with high turnover rates.

What are the Specific Causes of Employee Turnover?

The ADP analysis looked at 1,900 companies with 1,000 or more employees to identify attributes that can predict voluntary turnover. These “turnover drivers” have varying degrees of impact by industry and by company.

In aggregate, the category influences of turnover are ranked as follows:

  1. Pay
  2. Promotion
  3. Overtime and Premium Time
  4. Commute
  5. Experience and tenure

ADP RI Graph

The report also examines turnover probability by industry and the methodology for using benchmark data and predictive modeling to analyze the factors that contribute to voluntary turnover within an individual company. ADP suggests that application of predictive analytics at the company level can identify specific factors that contribute to the likelihood an employee will leave and spur the development of effective strategies to reduce the costs of turnover.

The complete report, Revelations from Workforce Turnover,  is available as a free download from ADP Reseach Institute.

Source:

1Revelations from Workforce Turnover, ADP Research Institute, p. 3

By | 2018-03-03T13:40:38+00:00 March 3rd, 2018|GEA Blog|Comments Off on Using Predictive Analytics to Reduce Employee Turnover
-