IBM Transforms Its Approach to Human Resources with AI

​IBM has spent the last five years using its artificial intelligence tools to design HR initiatives that evaluate employees and prepare them for internal mobility and career advancement.

The company is in the process of recreating its business from manufacturing computer chips and selling low-end hardware to investing in cloud computing, technology services and cognitive solutions, and wants to boost its talent pool to support that shift.

The importance of aligning the skills it will need to accomplish its goals to its business strategy can't be understated. IBM's workforce shrunk from 426,751 employees in 2010 to approximately 350,000 today. Yet whether it sheds employees or adds more (company officials can't say how many workers will be added when IBM finalizes its $33 billion acquisition of open-source software firm Red Hat), the focus is on improving employee expertise.

"Our biggest opportunity area as we look forward is growing skills and skills management," said Obed Louissaint, vice president of talent, Watson health & employee experience at IBM. Watson is IBM's technology suite of AI services, applications and tools.

To be successful, Louissaint said the HR department is partnering with business divisions across IBM to assess what skills will be needed, who has the expertise among the company's current pool of employees, how many employees with certain skills will be required and who needs to be trained to close the skills gap.

"We have modernized the way in which we think about jobs," Louissaint said. Since company executives believe AI and automation will change all jobs in some way in the next five to 10 years, the company is using the technology to harness information about IBM's workforce.

[SHRM members-only resource page: Skills Gap & Workforce Readiness]

Louissaint said approximately 50,000 IBM employees are enrolled in the company's Blue Matching program, which uses predictive analytics to generate a list of jobs currently available based on an employee's location, pay grade, job role, experience and other factors.

Once an employee opts in to the service, they receive weekly notifications to view potential job matches. So far 1,500 IBM employees have shifted to new jobs internally. 

Another AI-supported program called CogniPay helps managers make better compensation decisions by evaluating data on performance, what other employers pay for similar jobs, and what the demand is for similar skills. The analysis also includes internal forecasted demand for their expertise and voluntary attrition of employees with the same skills.

As a result of managers acting on CogniPay recommendations, IBM estimates attrition has been reduced by 50 percent. 

AI and IBM's Watson analytics are also being used to power Myca, the virtual career advisor behind the company's Watson Career Coach, an IBM program that helps employees identify skills gaps. Citizens Bank uses Career Coach for its own employees to suggest new jobs and recommend training, videos to watch and materials to read based on employees' career interests.

According to IBM's estimates, tools like CogniPay, Blue Matching and Career Coach saved the company more than $100 million last year alone. Total savings are calculated by tallying the expenses IBM avoided in sourcing and recruiting for new talent, hiring and training new employees, as well as the costs associated with the value of workers who would have likely left the company because they didn't envision career advancement there.

In fact, IBM's AI deployments have revealed the inadequacies of manager reviews, which are often subjective and don't give an accurate picture of the employee, Louissaint said. Because of this, IBM has scrapped the annual performance review in favor of assessing skills growth among employees on a quarterly basis. 

"We've moved from an annual performance rating to a coaching framework where we give employees feedback on an ongoing basis and an annual assessment on five dimensions, which are skills, business results, responsibility to others, client success and innovation. It's more about talent development than an annual appraisal," Louissaint said. 

In conversations with IBM's HR leaders, Ben Eubanks, SHRM-SCP, principal analyst at Lighthouse Research & Advisory, a human capital research and advisory services firm, said they told him they believe they're making progress with their use of AI tools. "They seem pretty happy with the way things are, and they believe AI pulls out a lot of the human bias that is inherent in HR decisions," he said.

He added that the company's recent workforce expansions and contractions have created a sharper focus on upskilling and reskilling workers. "One of the decisions they made a few years ago was that they had to be clear on those skills that mattered and that's one reason that their focus is on the skills conversation," Eubanks said. "If I have one piece of advice to IBM's team it would be to make sure that in the midst of the conversation around AI they focus on not losing the human piece. I feel like they are keeping a finger on that pulse."