Understanding employee retention using statistical and machine learning models
Abstract
Employee turnover can be very expensive for businesses. The company's overall performance is
impacted by employee retention problems. In the 21st century, having a strong mechanism for retaining
talented workers is emerging as a strategic advantage. Companies are beginning to understand that
their employees are their most valuable resource because it takes so much time and money to bring on
a new hire. According to one study, the cost of replacing a departing employee equals about 150% of
their annual salary.
Nowadays, organizations are trying to learn more about employees' intentions and the factors that
could lead to them leaving the company through a variety of HR analytics techniques. But nothing is
happening on a larger scale; everything is occurring in silos. The goal of this project was to provide
more clarity on the use case of machine learning in employee retention and to identify the critical
themes that cause employees to leave their current organization.
As part of the project, we conducted in-depth interviews, analyzed surveys, and conducted secondary
research. Our findings revealed that pay scale, location, career growth, and work quality are the most
important factors for employees. The career growth factor came into force when employees perceived
a lack of challenge in their current role and, as a result, looked out for new switch opportunities.
Furthermore, we discovered that employees' attachment to their work, team, and manager, as well as
the company as a whole, is critical for sustaining employee satisfaction.
Finally, in terms of the use case of machine learning in HRM, it has enormous potential and would
enable HRM functions to be more proactive rather than reactive. In comparison to other ML
algorithms and logistic regression, our report shows that the artificial neural network algorithm has the
highest accuracy for predicting employee turnover.
Our recommendation encourages businesses all over the world to perform the basic exercise of keeping
current employees happy, and machine learning algorithms will greatly assist in this endeavor. The
algorithm can predict which high-performing employees are more likely to leave. The HR department
can then take the necessary steps to reduce turnover. These measures would help to retain highly
productive employees while also ensuring their well-being.
Collections
- Student Projects [3208]