Algorithms are becoming increasingly relevant in the workplace. From sifting through resumes to deciding who gets a raise, many of these new systems are proving to be highly valuable. But perhaps their most impressive, and relevant, capability is predicting which employees will quit. IBM is in the process of patenting an algorithm that can supposedly predict flight risk with 95 percent accuracy. Given that we are in a candidate-driven market, this is a significant innovation. There are now more job openings in the U.S. than there are unemployed Americans.
Losing an employee can have a drastic effect on team morale, and result in a domino effect that leads to poor performance and productivity. Not to mention, it is expensive, and not just because of lost talent. It takes an average of 24 days to fill a job, costing employers up to $4,000 per hire — maybe more, depending on your industry. The good news is that only about a quarter of employees that leave do so within their first year. This means you have plenty of time to assess flight risks and address them.
But not every company has a fancy algorithm to help them out. Even predictive models that can identify the behavioral patterns that reveal who will quit don’t excel at explaining why they do. This is likely because the reasons people quit are deep-rooted and complex. During my 15 years working in data science, I have run countless predictive models on employee retention, student retention and customer churn across industry verticals, including healthcare, energy and higher education. Through my work, I’ve identified eight common leadership mistakes that help explain this why. Understanding them and how they impact your team will help you identify those who are at flight risk and make changes that may convince them to stay.
Please select this link to read the complete article from Harvard Business Review.