Solving Critical Workforce Issues With Predictive Analytics

Jason Feliciano recently joined JetBlue Airways as Senior Analyst, Workforce Intelligence, after transitioning from a major New York City hospital, where he served as a planning analyst for over two years. "Large organizations have their PeopleSoft application or an SAP that they use to manage the transactional data that they are collecting around the clock," he said. "They end up with a stockpile of data -- how many applicants, what job families, who are the hiring managers; terminations, hires, transfers, all that activity. But typically not much is done with it."

Q: How did you use HR metrics and predictive analytics to positively impact business decisions and prepare your organization to meet the challenges ahead?

A: At the hospital, we had the opportunity to work with that data and started to study turnover: trying to find out when are people most likely to leave, when are those pressure points in their tenure when they either stay or go. We didn't do much predictive analytics, this was more just wrapping our heads around what happens now. That was the foundation for the project that we worked on, which was trying to predict what our labor needs would be within the laboratory technology workforce in the coming years.

[eventPDF]

And then to piggyback on that, we used the turnover research that we'd done. How can we enhance our recruitment and retention capabilities. Revamp our recruitment strategies so we have a steady pipeline of talent to fill those positions that are probably going to be vacant. To do that research we acquired a PASW modeler, which is good to use when you have a lot of data. In addition to that, you have to make sure you have the infrastructure in place, because you need a consistent feed of data coming from multiple sources.

In organizations you've got repositories of data stored across the company but there's not much planning about how these repositories speak to each other. That's the big hurdle. We partnered with our business intelligence group and created a datamart that integrated our PeopleSoft data, our Kronos data, some of our Satisfaction data, employment data. You need to have consistent, clean data -- again, infrastructure is a big hurdle. Finally, of course, you have to get buy-in from the people in the organization who own this data. It's an enterprise-wide effort.

The purpose of it was to make a case for recommendations we presented to them. The laboratory workforce has been dwindling: for every two who enter the workforce, seven leave. There are only a third of the schools we had 30 years ago. The hospitals are all competing for the talent, and we're also competing with private laboratories. We expect to have an increased need for these laboratory workers. So we used that as a business case for developing relationships with the schools. We were able to use internal and external research, use the data we have to forecast what our needs will be in the future.

Q: What will you be doing at JetBlue?

A: It's different work, much more centered around tackling the issue of employee engagement. I'll be visiting some of the different airport sites. JetBlue has been collecting a lot of survey data for the last few years and they are looking to leverage it. We've got a variety of information from different points on the employee lifecycle and it would be good to see If there's any relationship with the amount of time they spend at the company and what their level of satisfaction is. There are different parts of engagement and we'll see if there are any relationships there.