IBM’s AI-backed ’employee retention’ software can predict when you’re going to quit with up to 95% accuracy

  • Software developed by IBM can predict when you’re fed up with your job
  • The tool is used and sold by the company to help retain employees
  • Important factors in attrition are long hours and fair compensation 
  • Machine-learning and AI are taking over the world of human resources 

By James Pero For

Published: 16:57 EDT, 8 April 2019 | Updated: 17:02 EDT, 8 April 2019

Hate your job? If you work at IBM or a company equipped with their artificially intelligent employee retention software, your planned two-week’s notice may already be old news. 

According to a recent panel discussion with IBM’s CEO, Ginni Rometty, at a Work Talent and Human Resources Summit in New York, the company’s ‘predictive attrition’ software is now 95 percent accurate in determining when an employee is ready to quit.  

Using their predicative software, Rometty says that IBM has been able to preempt employees on the cusp of quitting, bolstering their retention rates, which has reportedly saved them $300 million. 

Human resources may need a new name as a host of new software has begun to disrupt jobs and efficiency. Stock image


Among the many AI-powered tools being developed by IBM for human resources is ‘predictive attrition.’

The company says its analytical software can crunch data sets and predict which employees may leave the company.

From there, IBM says it can preempt an employees departure which saves the expense of training and on-boarding a new employee.

IBM’s CEO Ginni Rometty says that the software has saved her company $300 million.

‘The best time to get to an employee is before they go,’ said Rometty as reported CNBC.  

While the many ‘data points’ IBM’s program uses to predict when employees are most likely to quit are somewhat of a trade secret, a blog post from 2016 describing the use of its Watson Analytics — a smart data program — for the same purpose gives some insight into what may power the tool.  

In the post, the company’s analytics software takes into account a myriad of factors, the most important of which seems to be overtime hours worked — employees who worked 15 hours of overtime each week are at high-risk of quitting. 

Whether employees were compensated fairly for their extra work also play a major role, according to the analysis. 

Theoretically, if an employee shows signs that they’re ready to leave the company, management or human resources could make overtures in the form of added compensation or other substantive changes in order to prevent the attrition. 

According to CNBC, while the software has helped IBM retain employees and make human resources more efficient, it has also helped rid some employees of a job entirely — specifically, those in human resources.

Rometty said the software has lead to IBM reducing its human resources department by 30 percent globally.   

Interdicting employees about to quit isn’t the only function of IBM’s host of new human resource-based analytics and machine-learning software. 

IBM is now capable of judging both present and future performance within a 96 percent accuracy, according to the company. Stock image

The company has also rolled out performance-tracking programs that crunches data on employees’ projects, their skills, and weaknesses which they then use to dole out promotions and ‘increase efficiency.’

In a more paradigm-shifting twist, the company also uses Watson Analytics to judge not just an employees current performance, but how they might do in the future. 

Using its own algorithm, IBM says Watson can predict an employee’s future performance with a scary 96 percent accuracy. 

For IBM, the use of its human resources software isn’t just an internal affair. Analytics programs centered on employee management are also part of its business model. IBM currently sells a suite of tools to other companies that promise greater efficiency and, of course, its hallmark AI-driven prescience. 

For the average office worker, even those who aren’t yet planning on applying to IBM or the companies that it sells its software too, AI-backed human resources is a matter of fact. 

Companies are beginning to cut down their mound of job applications using AI-powered software that selects the ‘best’ resume based on user input and preference. 


Share or comment on this article:

IBM’s AI-backed ’employee retention’ software knows when you’re going to quit with 95% accuracy


via “IBM” – Google News

April 8, 2019 at 10:26PM