Solve the Turnover Mystery: Investigation Approaches Identified

Our last blog posed the question, “Can turnover risk be used as a proxy for actual turnover?” The short answer is yes. But keep in mind, the drivers differ. So, let’s look at how to use your employee survey to investigate turnover risk and voluntary turnover to identify the specific drivers at your organization.

Examining Drivers of Employee Commitment & Actual Turnover  

Your organization can examine how the various topics on the employee survey (management, com­munication, compensation, job fit) drive employee commitment. We refers to this as “Turnover Risk” – the likelihood that an employee may voluntarily exit the organization. For example, how an employee feels about job fit may be related to whether they would like to be working for the organization 3 years from now – both topics that may be covered in the employee survey. Next, organizations that use a confidential, identified survey can link back to employee survey data a few months after the survey and flag employees who have left the organization following the administration of the employee survey. For example, employees who voluntarily leave the organization can be “flagged” in the survey data for sub­sequent analysis of the employee experiences that related to whether or not an employee later left the organization. This identifies the key drivers of actual turnover among employees. These two approaches each have advantages and points to consider.

Investigating Turnover Risk

What to Do: To investigate turnover risk (commitment), the analysis can be conducted alongside any other survey analysis to understand right out of the gate, which employee experiences are critical in building, or conversely harming, employee commitment to the organization.

The Benefit: In SMD’s re­search across numerous clients, the company found that employee commitment, or turnover risk, is the strongest leading indicator of future voluntary turnover. When leaders can understand these key drivers alongside their survey results, they can create action plans to directly focus on building the employee experiences that will improve commitment and reduce the chances of future voluntary turnover.

The Cautions: The concern with this approach is that turnover risk is a complex phenomenon. While turnover risk (commitment) is the strongest leading indicator of future voluntary turnover among all the topics that are typically covered in an employee survey, it still does not account for a majority of the variance in actual turnover behavior. Meaning, while it may be the best predictor, there is a lot of room for false predictions. People who report being very committed still leave, and people that indicate on a survey that they are not at all committed often stay for a variety of personal or unavoidable reasons. Additionally, through SMD’s research, it found that the drivers of employee turnover risk are often differ­ent than the drivers of actual turnover. While this may seem to be a disconnect, think of it this way: the attitudes that influence other attitudes (experiences influencing felt commitment) can differ from the attitudes that influence subsequent behavior (experience influencing exiting an organization).

Investigating Voluntary Turnover

What to Do: When investigating voluntary turnover, revisit the employee survey and “flag” the employees who have taken the survey and have subsequently left the organization. This allows for an analysis of the experiences at work that are related to an employee’s voluntary turnover.

The Benefit: With this approach, the analysis directly links experiences to subsequent behavior, rather than an attitudinal outcome (i.e., commitment). SMD recommends taking this direct analytic approach.

The Cautions: This approach comes with a couple of requirements before it can be conducted.

  • Confidentiality is Key: The survey must be administered in a confidential, but identified way. In other words, respon­dents have to be able to be tracked at a later date so they can be coded as to whether they have turned over or not. Without a way to link employee attitudes to actual turnover, there is no way to use predictive analytics to determine the aspects of the employee experience that cause employees to ultimately leave the organization.
  • Patience: An organization must wait a certain amount of time after the survey to amass enough em­ployees who have voluntarily left the organization (who also took the employee survey) to be able to conduct the analysis. For some organizations that are relatively large or are experiencing high turnover rates, this may be as short as a few months. For others, it could take 6 months or more to have enough voluntary turnover to properly analyze the employee experiences that lead to actual turnover.