Ever heard of predictive succession planning? How about predictive career pathing or pervasive workforce analytics? Find out what predictive workforce analytics advancements are on the horizon, in this third blog post based on my interview with Greta Roberts, co-founder and CEO of Talent Analytics, Corp.

Greta will be a keynote speaker at The Predictive Analytics World for Workforce conference, which is being held in San Francisco from April 3rd through 6th in 2016. The conference is the premier workforce analytics event for HR professionals, business leaders, line of business managers, and analytics practitioners. This global, cross-industry event covers predictive solutions to today's greatest workforce challenges. Join Greta Roberts and APQC when you register today with 15% off code APQC15.

Elissa Tucker: At the Predictive Analytics World for Workforce conference, you will be delivering a keynote titled, “When the Robots Arrive: The Changing Landscape of Predictive Workforce Analytics.” Could you share what predictive workforce analytics advancements you see on the horizon?

Greta Roberts: At Talent Analytics, Corp., we spend a lot of time looking and learning from other more advanced predictive domains. These domains are about two years ahead of us, so by looking at them, we know what’s coming for the workforce space.

The first advancement that I will be talking about at the conference is actually already beginning to happen and we see momentum. It is to use predictions before you hire a person. The last year or so has been about predicting flight risk for current employees. Lots of people are excited about the possibility of having a flight risk score for everybody in their workforce.

The reason it’s important to predict before making a decision is demonstrated well with predicting borrowers that will default. Imagine if banks waited until after they had extended a loan to predict if the borrower was going to default or not. Silly, right? It’s the same when we predict, after we’ve already hired an employee, what their flight risk is. It is interesting to know but it doesn’t have a whole lot of value. So, the first advancement that I will talk about at the conference is predicting flight risk before you even hire someone — or predicting the probability of whether they will perform or not. This is the greatest ROI and it’s a wonderful place to begin with predictions.

Similarly, in predictive customer analytics, there was a mad scramble in the early days to ask: “Which of our current customers are going to leave?” They called it customer churn. Eventually they began to predict churn before they even engaged with the customer. This is exactly the same trend that we see happening now in the realm of predictive workforce analytics.

The second trend we see on the horizon is predictive career pathing and predictive succession planning. If you are going to be using predictions for your existing employees, this would be the perfect place to apply predictions. We’re doing this with our customers today. Imagine being able to predict for many, or all, of your employees, which would be the next best job for them, one that they would love, and one where they have a high probability of excelling.

The lovely thing about predictive analytics is that it can show opportunities the employees or the employer would have never considered, because our brains, as humans, continue to think that an employee should move up the same corporate ladder where we were hired. But predictive models are broader than that, and they’re less limited. All they care about is performance and success. They don’t care, and they’re not limited by the premise that if you started at the organization working in marketing, for example, you need to continue working in marketing.

Predictive models can help find a much broader pattern of success. For example, taking someone who is currently working in marketing and being very successful, and identifying that, wow, the same characteristics of the person working in marketing would give them a high probability of being successful as a data scientist. I just love how a predictive approach can broaden the opportunities for people.

Third, in terms of trends on the horizon—and this is a little further out—there is a concept of predictive models becoming pervasive. Pervasive means spread throughout and not special. Today, predictive models really are kind of special. If predictive models are operating in your organization, people are going to know where they are. They can point them out. They have to interact with them. As predictive modeling advances, we will find that the models are not nearly as special, and they’re much more just a part of our business and something we expect.

A good example of pervasive predictive analytics is in our cars. Many of us have anti-lock brakes. What these brakes do is prevent wheel lockup when your car is skidding. They predict that your wheels are about to lock up. And based on this prediction, they engage your anti-lock brakes. But how many of us have ever even thought: “There is a predictive model running here.”

Your car doesn’t actually show you a chart that says: “Hey driver, which of these options would you like me to do? Do you want me to keep you from skidding? Do you want me to do something else?” No, it doesn’t work like that, right? It’s pervasive. The predictive models are pervasive and they just work when you need them. So being pervasive is a natural evolution of the advancements that are happening. So we’re really excited about this evolution, as well.

Elissa Tucker: What else are you looking forward to learning about at the conference?

Greta Roberts: I think, like everyone else coming to the conference, I particularly look to learn about more ways people are using predictive to solve workforce challenges. All of us want to learn how other people are thinking creatively in applying these tools.

I’m particularly interested in one presentation where they are going to be able to predict the retirement of workers and the impact that is going to have. Predictions in the retirement area right now are really hot. There are a lot of people retiring and how can you predict that, and what is the impact going to be on the people, and also on the organization? That’s something I’m really interested in.

There are also two presentations on what the workforce domain can learn from predictive domains like the customer predictive domain or the risk predictive domain. I always love, and am very excited to learn from other domains that are more advanced. We all need to do this, to learn from other people that have gone before us. We can avoid their mistakes and it really helps with increasing the momentum for us.

And, I continue to be really excited about a predictive conference where data scientists who are working on workforce issues can gather together and share their best ideas and I just really can’t wait.

Learn more about analytics with the following APQC resources.

       Data and Analytics Terms (Infographic)
       Data and Analytics Glossary
       2016 Big Data and Analytics Challenges (Infographic)

Thanks for reading! @ElissaTucker