How to Predict Employee Turnover

Have you ever been blindsided by a top performer’s resignation? Omer Aziz, Chief HR Officer at FlightNetwork, may have cracked the code to predicting employee turnover. Inspired by solutions presented at Namely’s 2017 Redefine Live Hackathon, Aziz created his own method for determining when employees may be looking for a change. 

The Formula

Employees pursue other opportunities for a number of different reasons, but Aziz simplified the leading causes of employee fatigue into just five criteria. He determined that the most “at-risk” employees are those who:

1. Have not seen a title change in over a year

2. Have not seen a salary increase in over a year

3. Were highly rated during the last two performance reviews

4. Have a commute over one hour and fifteen minutes *

5. Experienced a life change in the past year

* Aziz advised that this is the “tipping point” in Toronto, Canada and urges HR professionals to determine the point at which a commute would become bothersome in their specific city. You can use home address postal codes to determine commute times.

High performing workers who haven’t experienced any recent changes in title, pay, or responsibilities may be looking elsewhere for a new challenge. Long commutes and big changes at home, like having kids, getting married or divorced, or having to care for a sick family member can also influence an employee’s decision to leave.

The Findings

Out of the 270 employees at Omer’s previous company, only 12 names met the first four criteria above. Aziz contacted all of the individuals’ managers, explained his methodology, and warned that their direct reports might be flight risks. Most of his warnings fell on deaf ears, but one manager took the advice to heart.

Five days after Aziz compiled his list, one of the at-risk individuals scheduled a meeting with his supervisor to discuss a pay raise. The manager, prepared with the heads-up from Aziz, was ready for the conversation and was able to negotiate a compensation increase for the employee. Content with his new salary, the employee continued to work for the company.

“No One Predicted Mike*”

Not all of Aziz’s findings led to happy endings. Aziz said one of his biggest surprises was finding out that one of the flagged names on his list belonged to a Managing Director, Mike. He was a dedicated executive leader who had been with the company for over 10 years.

Aziz called the CEO to tell him that Mike popped up on his list. The CEO was quick to dismiss his concerns. “We don’t have to worry about Mike. I talk to him every day and he’s doing just fine,” he said. Assured that Mike was content in his role, Aziz felt there was no need to worry. Mike handed in his resignation letter eight weeks later.

Mike’s commute was two and a half hours every day and he and his wife had just adopted two children. Predictably, he said that he needed to spend more time with his family and that his commute was taking its toll. He accepted an offer at a startup that was just a fifteen-minute walk from his house. 

The Conclusion

In the six weeks after Aziz had pulled his list, five individuals resigned and one negotiated a salary increase. Aziz had mixed feelings about the success of this formula. “Now, on the one hand, I was really proud I predicted this little train wreck, but on the other hand, I’m not very proud because these trainwrecks still happened. Next time I have to not only predict, but also take fast action.” Aziz encourages other HR professionals to run similar experiments by either using the data existing in their HRIS or pulling the data manually. However, Aziz encourages individuals to go beyond just predicting the future. Once you have an inkling of who might be unhappy, talk to employees to see how you can keep them engaged and take action to keep them around.

*Name changed.

Data equips HR teams with the information they need to make decisions and help their employees succeed. According to one report, over half of HR professionals hope to incorporate more data in their processes in 2018.