Data is the lifeblood of Experian, a global business that handles staggeringly large volumes of financial and consumer information. But when Barry Libenson arrived as CIO in June 2015, he found that the company didn’t distribute data between its divisions and to its customers as fast or as efficiently as it should. To address the situation, he launched several IT initiatives, such as implementing application programming interfaces (API), to improve data transfer, create other efficiencies and ultimately speed the pace of innovation.
Here Libenson shares his thinking behind APIs and some of his other strategic decisions.
Dossier: Experian CIO
Work-related factoid: Logged more than 250,000 miles last year traveling to corporate offices in Europe, Asia and Latin America.
Home base: Costa Mesa, Calif., where he lives with his two dachshunds
Biggest achievement: “I’m most proud of the number of technology leaders I’ve worked with over the years, and of mentoring top talent.”
If I wasn’t in IT, I’d be... a chef, specializing in Italian and Asian cuisine.
Hobbies: Cycling, scuba diving, cooking.
Best advice ever received: “I was having a heated discussion with someone I worked for, and he said, ‘If we agree all the time, one of us is redundant.’ One of the things I tell my team is I want contrarian views, because it makes us all better.”
Best time-management technique: “Getting to the office early, between 6:30 and 7 a.m., because I know I’ll get something done for at least the first hour of the day.”
What prompted your embrace of APIs and microservices? When I first joined Experian, I looked at the disparate data sources we have. We have data stored in different locations—[for example] DB2 or Oracle. It all depended on what part of the business you were in. Even within the enterprise it was difficult for the different lines of business to share that information, and information is one of the most valuable things we have.
So the first idea was to create an API extraction layer so it doesn’t matter what data form it’s in. You just make an API request for the information you want from the data source you want, and then there’s a translation layer that handles the aggregation and transfer of data. That allowed for easier data-sharing.
As we started to promote that concept internally, I started to hear that our customers have asked for the exact same thing. They want a microservice way of getting information from us rather than having to run an application to do that. It became very clear that it would be a critical component for how we delivered services internally and externally.
Were there risks in this approach? It’s easy to screw this up, and the ramifications for screwing it up were really bad.
Consider a financial institution that uses eight or nine of our different business lines. If each line of business built a different API interface independent of one another, with no collaboration, that would have been totally unacceptable for the customer. So governance was a critical component of this. Some of the lines of business had started down this path, but we put a freeze on that, centralized the activity, and put governance in place to avoid duplication and ensure a consistent behavior for customers. That hasn’t been without challenges, but it was the right way to go.
What was the biggest obstacle to moving your ideas forward? The stuff we’re talking about, this and other [initiatives] like building native hybrid cloud applications, it’s very leading edge. So finding developers and architects who have actually done it, that’s a whole new ballgame.
There’s a skill set evolving, but finding people who know [for example] how to build Hadoop applications, who know how to use containers, who are thinking about portability and moving applications, it’s hard finding those people. Everyone wants to hire them, so the market is very competitive.
How do you build teams and retain employees? When you find the people, you have to show them a good road map both personally and where the company is going. People are interested in learning and working on these new kinds of things, and they like a company that is embracing leading-edge technology.
I think Experian has done a good job in providing a compelling story as to why this is a great place to build these kinds of things. We’re doing cool stuff, and we have strong technology leadership across all the lines of business, which also helps attract talent. And we make sure our developers get exposed to our customers so they know why this is important and how impactful it can be.
What are the keys for leading a successful team? I very much believe that in the technology world today working collaboratively is really a key requirement to drive innovation. I’m a firm believer that great ideas come from all areas, and gone are the days when one person can know everything about one particular area. So I’m creating a very big tent and making sure everyone has an opportunity to make a contribution.
It doesn’t have to be a consensus-driven organization—at the end of the day, the debate needs to end and a decision has to be made—but it has to be an open forum where everyone can share their ideas and opinions.
What other major projects support your IT mission? My team delivered a series of services. The API framework was one of the services. Another example is our data fabric layer, which is one of the ways our businesses ingest information.
We get terabytes of information that we have to consume into our data stores. We don’t want everybody around the world doing that in a different way. It made sense to standardize that. So my team has built out the data fabric layer.
Our security platform is another one of those services. Another piece of all this is setting some design standards, getting everyone on the same page in terms of the programming languages we use and the rules governing hybrid cloud.
What was the biggest challenge you faced when you started at Experian? The real challenge for me was to get everyone moving in the same direction. The company had grown very rapidly, both organically and through acquisition. We operate all over the world. And the bigger you get, the more critical it is to find ways to collaborate and share more effectively.
And in order for us to speed up the pace of innovation and to deliver solutions more effectively, we focused on standardizing the technology platforms we were using and building out building blocks. I don’t want everybody inside the company building their own ingestion layers and security layers.
How did you get people on board with new initiatives? We’ve had a good journey in the past 18 months in terms of making that happen. Everyone is pretty much aligned around this and sees that they’re benefiting from the changes we’re making. The fundamental idea is that picking one version of Hadoop is better than having three different versions. It’s hard to argue with that single fact. So it’s, how do you identify what the right technology is? That’s where you can get wrapped around the axle.
But I’ve found that people are much more receptive to change when they’re part of the process than when you tell them what the change will be. I started to have conversations with the different teams. I tried to figure out which was the most pervasive [technologies] in the company and then it was a question of why others picked what they did, how do they feel about changing. It was making it a collaborative process and not trying to change too much too, too quickly.
What’s your next goal? I’m very much looking forward to the day when I can look at the Experian portfolio of applications and we’ve modernized the entire portfolio of externally facing applications and we can drive the pace at which we can innovate much more quickly. It’s difficult to do the second without having done the first.