Detailed Analysis of Turnover Can Help Employers Retain Top Talent, Consultants Say

Stay informed and ready to meet both everyday challenges and long-term planning and policy-making goals, with focused news, practical information, and strategic insights on all HR-related developments.


By Martin Berman-Gorvine

April 29 — Employers need to rethink the way they look at turnover by considering not only how many employees are leaving the organization, but also specifically which ones, Wells Fargo consultants said at an April 29 Conference Board webinar.

The traditional definition of employee turnover rate is the number of workers who left an organization in a given time period divided by the average number of employees. But while this formula is endorsed by the Society for Human Resource Management, it has two major problems, Camden Lee, vice president and senior analytics consultant in the Community Banking HR Insights & Analysis Group at Wells Fargo, said: it assumes that all jobs are the same, and that everyone who left the organization was performing at the same level.

That isn't the case. “People are not your most important asset—the right people are,” Leslie Golay, assistant vice president and senior analytics consultant in the Community Banking HR Insights & Analysis Group at Wells Fargo, commented during the webinar. “There's an amount of turnover that is healthy and necessary.”

Three-Tiered Model

For a more sophisticated analysis of turnover, Golay and Lee offered a three-tiered pyramidal model, building up from a “descriptive” base wherein one answers such questions as, “Was it regrettable? How does this compare to others? What was their performance rating? Where did they leave from? How quickly did they leave? How many left?”

The next layer is predictive: “Who is most at risk? What factors impact turnover the most? What happens if we decrease turnover? How does it impact staffing/recruiting? What if the trend continues?”

At the peak is a prescriptive layer: “What policies promote our best people to stay? What is the optimal amount?”

Some Turnover More Costly

Measuring the wrong thing can provide incentives for the wrong behavior, Lee warned. Thus, the loss of “high-impact talent goes unnoticed,” as does the fact that the employees who don't leave may be underperforming. If managers are rated the same on their retention rate, regardless of whether they lose high- or low-performing employees, “you may be keeping the wrong people,” he said.

In reality, a group that loses a greater proportion of high-skilled employees loses intellectual property, faces higher costs in replacing them and faces a long “ramp-up” time for their replacements, Golay noted. So if there are two groups of 100 employees each, and they both lose 10 employees, but in the first group only two of the departing employees were highly skilled while in the second group five were, it's the second group that has bigger problems, she indicated.

Measuring Employee Value

Building a weighted turnover model requires first building a model of each employee's worth to the company, which in the Wells Fargo schema starts with “objective” factors like market value and production performance, but including “subjective” factors Golay detailed, noting that some employees are high performers “although they are only average at whatever our measuring widget is.” Thus, performance reviews, recognition awards and whether an employee is viewed as “high potential” need to be taken into account, along with the person's “fit” within the organization and other factors such as whether he or she contributes to desired organizational diversity.

Golay said the weighted individual scores are useful in many more areas than just turnover. On the individual level, they can be used to identify high-potential employees and standardize what that term means across different regions, compare employees, develop succession planning, and contribute to recognition programs or coaching opportunities. On a group level, the individual scores can be used “to identify leaders who are really good at retaining talent,” she said.

Lee and Golay also said the scores can be used for risk modeling of turnover, “understand[ing] the likelihood and severity of impact to the organization if turnover occurs in key groups” and “determin[ing] groups that may need intervention.” Managers who are losing key talent can be retrained, they added.

To contact the reporter on this story: Martin Berman-Gorvine in Washington at

To contact the editor responsible for this story: Simon Nadel at