Employee attrition analytics is specifically focused on identifying why employees voluntarily leave, what might have prevented them from leaving, and how we can use data to predict attrition risk.
Employee attrition analytics is specifically focused on identifying why employees voluntarily leave, what might have prevented them from leaving, and how we can use data to predict attrition risk. Most importantly, this type of employee predictive analytics can be used to help organizations understand and design the interventions that will be most effective in reducing unwanted attrition.
Over the past two years, this type of analytic practice has become indispensable. Global labor markets have swung dramatically due to the COVID-19 pandemic, and in August 2021, 55% of the American workforce said that they plan on looking for new employment over the next 12 months.
In addressing the ongoing challenges of the pandemic and the rise of remote work, employee attrition analytics will remain important to organizations seeking to retain top talent. Predictive analytics capability enables the design of an employee retention model to keep these valuable employees engaged and on board.
o predict future patterns, we first look to the past to answer the who, when, and why questions. As we noted in a previous article, we can find the answers to these questions by using engagement survey data collected six months to one year in the past, and creating a post-hoc demographic of employees who left the organization voluntarily. Analyzing this demographic will reveal information about turnover in various job roles, tenure levels, business units, and locations – and reveal pockets of high turnover – to tell us who is leaving and when.
An employee listening perspective will answer the question of why. We can look at what employees who left were telling us about the workplace, work relationships, and their sense of connection to the organization in the months before they left. The comparison of engagement survey data to termination data can reveal areas of the employee experience in need of improvement. We can also look at how the responses of employees who left the organization varied from those who stayed to see which factors in the experience might have been barriers to engagement. This method can be used by any organization that conducts engagement surveys and has the ability to group employees by various demographic factors.
Exit surveys are another potential data source that can provide richer information. Comparing responses on exit surveys to employees’ engagement survey responses can reveal how the employees’ perceptions changed over time. Correlating exit and engagement survey data can yield additional capability to predict attrition risk.
Answering the who, what, and why questions – and combing the data to see other similarities and differences between employees who stayed versus those who left – is the foundation of employee attrition analytics. An effective employee retention model must be built on the solid footing of the data; otherwise, actions intended to impact attrition are at best only guessing at how to solve the problem – and may be guessing at where the problem actually lies.
Uncover the factors that lead to employee attrition and explore important questions such as ‘show me a breakdown of distance from home by job role and attrition’ or ‘compare average monthly income by education and attrition’. This is a fictional data set created by IBM data scientists.
Education
1 'Below College' 2 'College' 3 'Bachelor' 4 'Master' 5 'Doctor'
EnvironmentSatisfaction
1 'Low' 2 'Medium' 3 'High' 4 'Very High'
JobInvolvement
1 'Low' 2 'Medium' 3 'High' 4 'Very High'
JobSatisfaction
1 'Low' 2 'Medium' 3 'High' 4 'Very High'
PerformanceRating
1 'Low' 2 'Good' 3 'Excellent' 4 'Outstanding'
RelationshipSatisfaction
1 'Low' 2 'Medium' 3 'High' 4 'Very High'
WorkLifeBalance
1 'Bad' 2 'Good' 3 'Better' 4 'Best'