What is the best way to hire top-quality data scientists? And given that they are so often in short supply, how do you prevent the superstar you hired from being poached by other companies?

Data-analytics experts offered advice during a recent session at the Kellogg School of Management’s Growth Forum. Below is an edited excerpt of their discussion about building an analytics team, led by marketing professor Eric Anderson.

ERIC ANDERSON:

Where do you find talent? What are you looking for?

DAN WAGNER: CEO and founder of Civis Analytics, formerly chief analytics officer for the Obama 2012 campaign

I’ll tell you the story of the smartest person I ever hired. I said, “Why do you want to work for me?” He goes, “I don’t know. I just think it would be fun or something.” This guy ended up being the top person that I ever hired, a complete genius. But he was not capable of expressing his thoughts in words. Now, in most companies, he would get screened out. What I’ve learned from the thousands of people that I’ve interviewed is I don’t have the intellectual capability of having the conversation with a human being and assessing whether or not they’re going to be good at this job. What you can do is simulate the job for them through an exam process. See how they do and rank them against everybody else. What you’ll find is that the introverts tend to do much better, and that is the classic person that you’re trying to hire. There is way less pedigree bias in that. There’s less, “I went to University of Chicago, I want these people. I went to Harvard, I want these people.” That’s stupid. There’s less gender bias in that. People have things going on in the back of their brain that you can’t control. You want to make sure you have an objective measurement of intellectual capacity.

LESLIE HAMPEL: Director of global strategies at Starbucks

What I am looking for is someone who can bridge the gap. Can you do the math? That’s important, but can you pull the story out of the math? Particularly for the next 10 years or so, as we work our way through this current generation of CEOs who don’t understand algorithms for the most part. They are making decisions from a very different place. How do you show them that you have applied the analytic rigor, then help tell the story so that they feel comfortable investing millions, if not billions, of dollars in this idea? It really becomes about storytelling. Finding someone who can do both coming straight out of school is almost impossible. I look for people who are curious. I need you to be so curious that you’re going to keep digging into that data until you find the nugget of insight that really unlocks the whole story.

SCOTT JONES: Vice president of mobile applications, analytics and personalization at Nordstrom

I’ve not found any predictability in terms of where really good, high-quality, and yet pragmatic people come from. We’ve had life sciences folks come on board. We’ve had someone from automotive engineering come on board. They hook on really quickly to how to sell fashion handbags. Don’t allow appearances or an expectation on background to drive it. As you find that great team of people, it is your job to make sure you’ve got bigger and more challenging problems to throw at them. Make it meaningful. People have to have a sense that, you know what, this company is understanding more, getting into the customer experience more, and making decisions based on what I’m doing. If you don’t do that, the best ones, they have 100 opportunities out there. Every single day, people are calling them.

"Whether people like it or not, the robots are coming and sometimes they wear glasses."—Dan Wagner

ANDERSON:

How do you address the question of what’s my future? When they look up the ladder, they don’t necessarily see people that look like them today.

JONES:

I think you have to be very candid and honest with people and say, “We’d love to keep you here for the next 20 years and throw all kinds of stuff at you.” At a minimum, you’re handing them the ability to do meaningful work in this emerging space. Two or three years down the road, as the company evolves, we’d love to throw bigger and more exciting challenges [at them]. Some of these roles, as the company becomes more data driven, some of those roles will mutate. But at a minimum, I want to make sure that we help them make meaningful career progress and that they have great options down the road. I want to make sure that we don’t leave someone in the corner with no options when the time comes.

HAMPEL:

We have to convince people to create new [career] pathways. And part of that is by putting your head down, working really hard, and proving your value, so that two, three years down the road when you’re ready for that next step, someone’s willing to invest in you.

WAGNER:

If you are able to understand, create conclusions, and communicate those conclusions to a wide group of individuals, you have more leverage than pretty much anybody else in the organization. I think a lot of them, frankly, in 10 years are going to be CEOs. I have high faith in them. I think about that moment where Mitt Romney was sitting with his advisors at noon on Election Day and they were saying, “Things are so great. We’re going to win. Everything’s great. Look at this poll thingamajiggy.” Three hours later, the anxiety started to go up and then five hours later, they were the laughing stock of the planet. People have a choice, organizations have a choice: Are they going to be Mitt Romney? Organizations that decide to live like Mitt Romney are not going to exist in 10 years. Whether people like it or not, the robots are coming and sometimes they wear glasses.

For more insights on hiring and retaining data scientists, read our article on the topic with marketing professor Eric Leininger.