There is a new tech in town for predicting when you will be resigning. DBS has developed a system which uses predictive algorithm for determining when an employee will be quitting.
It takes into consideration about 600 data points which include, absenteeism, salary increase, rate of promotion, the birth of a child, and training modules attended.
All of DBS’s 11,000 employees are under this system. It will churn out a monthly report which will indicate the predictions of the likelihood an employee is going to resign.
It will also indicate the following steps to be taken by managers if they are anticipating an employee to be leaving.
Banks in Singapore experience high turnover rates, up to 30 per cent, mostly in the sales department.
Hence, this new system will allow human resources to be better prepared when such a situation arises or even prevent them from happening.
Ms Lee Yan Hong, managing director and head of group human resources at DBS Bank said that employees sometimes leave due to family reasons, which is out of the bank’s control.
But is also sometimes is due to reasons like employees feeling that they are being underpaid or undervalued. For such cases, the bank would want to reach out to the employee and find out why they are feeling as such and provide them with the required assistance.
“Our focus is to make sure that we do not lose people for the wrong reasons,” she said.
Ms Lee said that the system was developed through data gathering and the formulation of the algorithm by four data scientists and three data analysts from their human resource department.
The bank is now prioritizing change management. Managers will have to take actions based on the results churned up in the monthly reports.
DBS has shared that it estimates saving S$5 million worth of spending in areas such as recruitment, training, and operational downtime if it can reduce its annual turnover rate of 15 per cent bank-wide to just 1 per cent.
This is not the first time that DBS has engaged in tech and innovations for improving its services and management. In an earlier OpenGov earlier article, we reported about the way in which DBS had transitioned into a data-driven organisation.
DBS had built a central data team and enterprise data hub which allowed the bank to grow economically and having the ability to experiment more.
As such, DBS can use this hub to store and analyse a multitude of events and be better able to formulate answers for questions before they are being asked.
This allows for a better engagement of customers and allows the bank to provide better service to its customers.
The employment of machine learning of the call logs, received by the bank’s call centre, will allow for customer’s concerns to be understood and for the bank to take immediate action on it.
DBS’s staff has been better able to engage in more experiments and be more involved in innovations surrounding the enhancing of customer service.
