“For all of my career, I've been trying to catch people after they do something horrible. For once in my life, I'd like to catch somebody BEFORE they do something horrible, all right? Can you understand that?” - ATF Agent Doug Carlin (Déjà Vu, 2006)
One of my favorite time travel movies is Déjà Vu from 2006. In it, Denzel Washington plays ATF Agent Doug Carlin. He works with a team of scientists and travels back into the past to prevent a disaster from happening on a ferry in New Orleans. I think we all would love to travel back in time to stop something bad from happening. In business, losing a key employee when you could have prevented it can be tragic as well. It can cost, on average, some $3,341 to hire a new employee, according to the Society for Human Resource Management. This is why companies are investing in predictive analytics to help salvage their employees.
According to an article from Entrepreneur Magazine, in the third quarter of 2014, 47% of employees felt confident that they would find a new position in the next six months that would be fitting for their experience level. This is up 11 percentage points from four years ago. In addition to a strong job search market, there are some clear behavioral signs that employees may be looking:
Coming in chronically late or early
Becoming less friendly / more agitated with colleagues / employee engagement
Completing a new degree / certification
A major life event, such as divorce or marriage
Significant company restructuring / changes
With all these predictors, it’s no wonder that data scientists are now creating algorithms to help companies turn these warning signs into a true predictive science. In March 2015, The Wall Street Journal reported that data scientists are creating models to help predict which workers might leave first.They refine variables depending on which are the most predictive given a specific company or specific group in an organization. According to the article, “VoloMetrix Inc., which examines HR data as well as anonymized employee email and calendar data, found that it could predict flight risk up to a year in advance for employees who were spending less time interacting with certain colleagues or attending events beyond required meetings. And Ultimate Software found a correlation between a client’s employees who waived their benefits coverage and those who left the company.”
Keep in mind that predictors are fine and dandy, but, without action on the part of the company, they are useless. If it’s found that John Doe is slated to resign in 6 months if there is no opportunity for advancement, then the company needs to find a way to advance John within the company. It’s not easy; there are times when a company cannot help their employees because there are genuinely no opportunities. Communication between managers and their direct reports becomes very critical.
We may not be able to go back in time like Denzel Washington, but a combination of predictive analytics and true organizational leadership may be the step in the right direction to save your employees.