Drive Turnover Down

Data reveal causes and patterns that help you enhance retention.

July 1, 2012 | Adrienne Fox

Employers haven't had to worry much about turnover during the past four years—two years of a Great Recession followed by two years of a jobless, anemic economic recovery. Voluntary turnover rates decrease as unemployment rises. In January 2009, when the unemployment rate passed 7.5 percent, the number of nonfarm employees voluntarily quitting their jobs sank to 2 million from 3.5 million in January 2001, when the unemployment rate was just above 4 percent, according to the U.S. Bureau of Labor Statistics.

The quit rate has remained low since 2009 as employees hunkered down, many thankful to have jobs. Executives "say they're not worried about turnover right now because their turnover is low … because there's nowhere else for people to go," explains David Allen, SPHR, professor of management at the University of Memphis.

That's beginning to change. In May, the U.S. unemployment rate stood at 8.2 percent, a decrease from a year earlier. As the labor market opens up, HR professionals may face a turnover crisis.

The length and depth of this economic downturn has researchers predicting massive turnover in the near future. A whopping 84 percent of employees said they planned to search for a new job in 2012, according to a November 2011 online survey of more than 1,000 people in the United States and Canada by Right Management.

The results of an October 2011 Mercer survey were significant but not as startling. The percentage of workers seriously considering leaving their organizations jumped from 23 percent in 2005 to 32 percent in 2010, according to the survey of nearly 30,000 workers in 17 geographic markets, including more than 2,400 U.S. workers.

Already, 34 percent of HR and hiring managers have reported that voluntary turnover at their organizations rose in 2011, according to CareerBuilder's 2012 U.S. Job Forecast report, which was based on a survey conducted by Harris Interactive. The desire for higher compensation and feeling overworked were the top reasons employees gave for resigning in the survey of more than 3,000 hiring managers and HR professionals across industries and company sizes. Thirty percent of employers said they lost top performers in 2011, and 43 percent said they were concerned that top talent could jump ship in 2012.

"We have pent-up turnover because of the jobless economy," Allen says. HR professionals have "to work now to tweak the employee experience so that they will stay and be engaged. You need to know who is at risk and how to retain them."

Tracking, dissecting and predicting turnover can keep your organization from falling victim to the predicted wave of exiting employees.

Turnover 101

To prevent turnover, first know what it is. In basic terms, turnover measures the number of employees who have left an organization in relation to the number who have stayed. HR professionals track both voluntary turnover (when an employee chooses to leave) and involuntary turnover (when the employer initiates the separation.)

While HR leaders may focus more on preventing voluntary turnover, involuntary turnover data also are valuable, according to Karen O'Leonard, director of research methods and analytics at Bersin & Associates, an HR research firm in Oakland, Calif. "Involuntary data may reveal issues in recruiting methods or development plans or performance improvement interventions."

According to the Society for Human Resource Management's (SHRM) 2011-2012 Human Capital Benchmarking Database, the industry with the highest turnover was services—accommodation and food and drinking establishments. Its average rate was 35 percent. At the low end were association and utilities industries, tied at 8 percent. Across all industries, average turnover was 15 percent.

Only half of 725 companies of various sizes tracked turnover of top performers, according to Bersin & Associates.

Examining an industry's aggregate numbers is a good way for HR professionals to benchmark how their companies are doing, but the aggregate data offer only a starting point.

"I look at benchmarks to see if I am in the ballpark, and then I have to determine what is high-performing in my organization," says Cathy Missildine-Martin, SPHR, chief performance officer at Intellectual Capital Consulting in Atlanta. "Look at your historical data on turnover. How has it changed?"

"Too many times, we see a turnover rate and we make plans to reduce the rate by an arbitrary percent," says Dwane Lay, director of human resources at Missouri Baptist Medical Center in St. Louis, which has 3,000 employees. "Executives will say, 'We need to cut turnover.' Well, by how much? What number is low enough? Some turnover is healthy because it weeds out the disengaged and brings in new people with fresh ideas. What number is too high?"

High turnover can be expensive: The cost to replace and hire new staff is estimated to be 60 percent of an employee's annual salary, states a 2008 SHRM Foundation report. But total costs of replacement, including training and loss of productivity, can range from 90 percent to 200 percent of an employee's annual salary, according to a 2006 PricewaterhouseCoopers white paper.

Industries with lower turnover, such as utilities and professional trade associations, report higher revenue per full-time equivalent than industries with higher turnover, such as services and recreation, according to the SHRM database.

"We all talk about turnover, but we don't care about turnover," Lay notes. "We only care about what portion of it is having a negative impact on the bottom line. To find out, you have to get behind the number."

Turnover 2.0

Comparing good turnover vs. bad turnover or preventable turnover vs. unavoidable turnover are ways to dissect the overall number of people leaving an organization each year. Turnover should be segmented "just like marketing segments their markets and customers," Missildine-Martin says. Take the case of a manager who develops his employees for internal promotions. That results in positive turnover that should be captured, she explains. Turnover resulting from external circumstances, such as relocations or retirements, should be separated out of the aggregate.

The Boys & Girls Clubs of America in Atlanta currently has 370 corporate employees. Terri Dorsey, director of organizational development, tracks voluntary turnover and separates out retirees and those who leave headquarters to work at a Boys & Girls Club.

"Most of our turnover is tenure-related and occurs among people who have been here one to three years," Dorsey says. In exit interviews, she has determined that the reason is either a bad fit or the job wasn't what the person expected.

As a result of that data, last year the HR department implemented a more "high-touch" recruiting system that brings in candidates for multiple interviews on-site. Recruiters changed how they measure their results to reward finding a good candidate rather than any candidate. "We used to look at time-to-fill metrics, and now we're tracking quality of hire," Dorsey says. "We ask the supervisors to evaluate the quality of hire, and we look at performance reviews after the first year."

Among the 82 new hires for 2010 and 2011, Boys & Girls Clubs retained 93 percent, she says.

The turnover rate dropped slightly in 2011 to 9 percent, down from 11 percent the year before. "That number would scare us if it applied to our high-performers," Dorsey says. "We are in the upper 90s for retention rate for high-performers."

She tracks the number of top performers and high-potentials who leave the organization—metrics not all HR professionals are able to produce.

Only half of 725 companies of various sizes track turnover of top performers, according to Bersin & Associates' Talent Management Factbook 2010.

When Nancy Nygren took the position of chief executive officer of the Girl Scouts of Orange County in California five years ago, she saw opportunities to reach more girls. New goals and objectives required fundamental change in the company's culture, and that led to higher turnover among resistant employees.

She says such turnover was healthy and, since 2010, staffing has stabilized at 65 employees. Executives continue to monitor responses to questions on an engagement survey. From 2009 to 2011, the proportion of employees who said "I want to stay at Girl Scouts of Orange County for more than a year" increased from 57 percent to 72 percent. The percentage who said "Girl Scouts of Orange County motivates me to give my very best work" jumped from 63 percent to 76 percent.

Nygren wants her HR professionals to be able to tell her who is at risk of leaving and whether those individuals are top performers.

"The C-suite doesn't want to know the organization has 60 percent or 5 percent turnover," Missildine-Martin says. "They want to know: Of those 60 percent or 5 percent, how many are top performers or high-potentials? And is it impacting the bottom line?"

Capturing why an employee leaves is important but tricky, Lay says. Every human resource information system "has termination codes to input why someone is no longer working there," he says. "We allow managers to tell us what the termination code is. If the reason someone is leaving has anything to do with their manager, then you won't get an accurate picture."

Exit interviews, conducted by an HR professional or a third party, have supporters and detractors. They are usually conducted within a week of the separation, but Allen says research has shown a difference in the answers that an exiting employee gives one week after leaving vs. three months later. "The separation, even voluntary, is too emotionally charged to allow for insightful and accurate answers right after leaving."

Take emotion out of the equation and analyze the data for patterns that reveal reasons behind turnover, Lay suggests. "If we know that turnover jumps 10 percent when we promote someone as a first-time manager, then we can beef up our new-manager training," he says. "If we know turnover spikes seasonally or after annual bonus payouts, then we can restructure bonuses to spread them out during the year. If there is a jump in turnover after placing a high number of new people in one department, then we can work on team building or onboarding."

Predicting Turnover

Predictive analytics represents the next wave of HR metrics, and it makes sense to use them to forecast turnover. "When you report on turnover, you've missed the opportunity," Lay explains.

Using turnover data with other data to predict future behavior and to intervene moves such analyses from the passive to the proactive realm. "Some progressive HR departments have hired people with organizational development or statistical backgrounds to do predictive modeling," O'Leonard says.

Even when organizations don't invest in analyses to reveal attrition variables, HR professionals can uncover risks associated with turnover, says Brian Wilkerson, principal consultant at Revolution Advisors in Louisville, Colo. Wilkerson has identified general attributes and behaviors that he says affect turnover at any organization. They include:

The local economy.
The traits of the job.
How often employees have been promoted.
The pay increase frequency.

For instance, departments that have been consistently staffed under peak levels are at risk for high turnover due to built-up resentment and burnout, he explains. "As you start hiring again, look at departments with the largest gap between current staffing level and peak staffing level and fill those holes first. We have a tendency to pay attention to the hiring manager screaming the loudest or the easiest job to fill."

Promotion statistics can be a harbinger for turnover among high-potentials. After they've spent three years in the same position, "start to worry about your top performers," Wilkerson advises. "That number is longer because of the recession, but it will go back to around three years once the labor market opens again."

Each company tends to develop a "break point," the length of tenure when employees are likely to stay for the long haul. For example, it may be that employees who make it past the four-year mark at your company stay for another six years. Find out what your break point is, Wilkerson suggests.

The Bureau of Labor Statistics measures six types of unemployment. The broadest measure, known as the U6 unemployment rate, includes those who have given up looking for work and the underemployed. These data, when taken from your local area, can be a good predictor of turnover. Wilkerson's research found that when the U6 rate increased by 0.5 percent, turnover decreased by 1 percent to 4 percent at companies studied. Also, when personal consumption goes up, turnover increases 30 to 60 days later as the labor market expands to meet increased demand, Wilkerson has found.

So, in one company, turnover could best be predicted by looking at a combination of the employee's time with the organization, the U6 unemployment rate, the employee's most recent assignment, and the history of bonuses and promotions. In another organization, turnover was based on demographics, gross domestic product growth and time with the company. In many cases, factors varied by position.

Wilkerson pinpoints certain employee behaviors that are signs of intent to quit. Hours worked and workload are predictors, when adjusted for the organization's culture. "Working longer or working less—a change in work patterns—is indicative of desire to leave," as are "doing less collaborative work such as volunteering for special projects or asking [fewer] questions," he says. "If someone gets quieter than they normally are, that's the biggest behavioral cue of attrition risk. It could be a sign of a problem unrelated to work, but either way it's an opportunity for the manager to sit down with the employee and find out what's going on."

Wilkerson says HR professionals can train managers to look for signs and in how to have these conversations with employees.

Burnout remains a key indicator for future turnover, especially among nonprofit organizations, notes Lynne Norton, director of strategic partnerships at the Georgia Center for Nonprofits in Atlanta. "Nonprofit employees are similar to first-responder workers who are vulnerable to emotional burnout."

At Girl Scouts of Orange County, Nygren relies on her HR team to get a sense of employee burnout. "Our HR department is the eyes and ears of the staff, who feel very comfortable confiding in HR," she says. "We, like many organizations, are stressed and understaffed. When we start noticing individuals at risk of leaving because of the stress, we intervene."

HR leaders need to make the case that it’s worth investing in retention right now—even if turnover is low.

Wilkerson encourages HR professionals to use "pulse" surveys more often to uncover attrition risks. However, such surveys "don't replace yearly, companywide engagement surveys."

Putting It Together

For business leaders, the data are all there—turnover, employee engagement, recruiting metrics, performance evaluations. What's needed is someone to slice and dice the data, look for patterns and signs, and intervene to prevent undesirable turnover.

What's needed "is an HR professional who has an interest in analytics and knows what questions to ask," Lay says. "It doesn't require a doctorate in statistics," although he recommends taking classes.

William Wolf, managing director and global head of talent development at Credit Suisse in New York, advises HR professionals wanting to use predictive analysis to "walk before you run." First, make sure the data collection is streamlined and not duplicative. Executives must have confidence in the quality of the data before HR professionals can predict turnover. He suggests looking for people in marketing, finance, risk or the front office to apply analytical skills to people data. "Since we started to share the team's findings from the attrition analysis, people have come out of the woodwork wanting to work in HR," Wolf adds.

Once the right team and data are in place, discern the factors common in those who stay and those who leave.

After you have identified factors that predict turnover in your organization, monitor them frequently, Wilkerson advises. HR leaders need to make the case that it's worth investing in retention right now—even if turnover is low.

Wolf recommends investing in predictive analysis because it relies on objective facts from employees' profiles rather than what someone chooses to tell you in an exit interview or traditional assumptions, such as: Jane left because she's a woman, so there must be a lack of opportunity for women.

"Predictive analysis uncovers the real, underlying reasons people stay or leave," summarizes Wolf. "Investing in people analytics in this way will bring real credibility to the HR profession."

The author, a contributing editor and former managing editor of HR Magazine, is based in Alexandria, Va.

What insights has your organization gained from its turnover metrics?