The new hire experience should be evaluated and considered a strategic tool to uncover the key factors that drive turnover in an organization. For example, the misalignment between an employee’s expectations and reality once on the job can oftentimes be a root cause of turnover. Here’s how to use data to discover the company downfalls that lead to turnover.
The integration of onboarding data with other business outcome data to show prioritization and impact is possible with smarter analytics by integrating two key data elements for the analysis — the onboarding survey data (i.e., 30-day onboarding survey and 90-day onboarding survey) and actual turnover data.
Step 1: Identified Survey
To conduct this analysis, leaders must take a crucial step prior to survey administration and conducting an identified survey (i.e., track survey responses to an individual employee). Without an identified survey (i.e., no unique identifier) or any employee information captured in the survey that enables matching the two data sets together, leaders won’t be able to recognize whether the survey-taker has left the organization or is still employed.
Step 2: Multiple Logistic Regression Analysis
Next, statistical software will be necessary to conduct a multiple logistic regression analysis to regress the turnover from the HRIS onto, first, the 30-day onboarding survey categories. This analysis is important because it will tell exactly which survey categories have the biggest impact on an employee’s decision to exit the organization. This analysis tells leaders why their employees are leaving the specific organization. Smarter analytics allow for conducting a valid diagnosis of what contributes to employees leaving the organization as early as their 30-day anniversary. The same approach is used to analyze the 90-day onboarding survey.
Step 3: Focus Areas for Reducing Turnover
The output of this smarter analytics approach is a brief list of the most-important areas (or key drivers) to focus on in recruiting, hiring and onboarding processes that have the most significant impact on reducing turnover in a given organization. This list tells a different story than when an organization simply tracks scores or looks at high and low scores. The story now is, “If we improve elements X, Y and Z, then we can reduce turnover by 4 percent, which carries a savings of $2 million.” Without smarter analytics, the only story to be told is, “Our onboarding survey results went up a little bit this quarter and went down a little bit from two quarters ago.” That second story does not carry much weight, does it?
Findings: Which Categories to Act On?
The next step is to use logistic regression to determine which experiences were the strongest predictors. Basically, this approach allows an organization to understand which areas of the 30-day onboarding (or 90-day onboarding) survey truly differentiated the reasons why those who voluntarily exited made their decision and why those who stayed with the organization did not. In other words, this approach revealed the “why” of turnover, which is critical to limiting turnover in the future.
Matt Betts is the director of product development at Strategic Management Decisions. To comment, email firstname.lastname@example.org.