Machine Learning Times
Machine Learning Times

2 years ago
Wise Practitioner – Predictive Analytics Interview Series: Greg Zeeman and Joe DeCosmo at Enova International


In anticipation of their upcoming conference co-presentation, How Digital Decisioning Can Help Your Business Adapt Quickly to Regulatory Changes, at Predictive Analytics World for Financial in Las Vegas, June 3-7, 2018, we asked Greg Zeeman, Chief Operating Officer, and Joe DeCosmo, Chief Analytics Officer, both at Enova International, a few questions about their work in predictive analytics.

Q: In your work with predictive analytics, what behavior or outcome do your models predict?

A: As a global online lending and financial technology company with ten brands in three countries, we use advanced and predictive analytics to guide virtually every decision we make with Enova’s more than five million customers. Simply put, our products and brands are driven by our ability to accurately predict customer behavior and value. We are most often predicting credit risk — the risk of loan loss due to fraud, credit worthiness, or affordability. We use dozens of data sources, tens of millions of customer transactions, and a suite of machine learning and regression-based models to make these predictions as accurate as possible. This all happens in real time and in sub-seconds. In this way, we can ensure a fast and easy customer experience and make the best lending decisions.

Q: How does predictive analytics deliver value at your organization – what is one specific way in which it actively drives decisions or operations?

A: Predictive analytics is a core component and competitive advantage for Enova that drives value across the whole organization. As Greg Zeeman, Enova’s COO, and I will talk about at PAW, we have used predictive analytics in surprising areas like compliance. When we needed to quickly adjust to comply with a 2015 regulatory change regarding the use of automated, electronic payments from customers, we immediately got to work and built a system of models and optimization routines that enabled us to make the necessary changes on a tight timeline while ensuring a good customer experience and profitability.

Q: Can you describe a quantitative result, such as the predictive lift of your model or the ROI of an analytics initiative?

A: While we drive ROI every day with our predictive models, one noteworthy example is the custom credit model we built for our fastest-growing US brand, NetCredit. When we launched NetCredit in 2013, we relied on existing credit scores and reports to make our lending decisions. Once we had enough data, we built a custom credit model and decision process that continues to outperform existing credit scores by more than 50%. This huge lift in predictive power has enabled us to grow NetCredit to be our largest loan portfolio and successfully extend credit to a much wider range of customers than traditional lenders. We are now on our 8th generation of this credit model and continue to drive significant performance improvements with each new version.

Q: What surprising discovery or insight have you unearthed in your data?

A: One surprising discovery we made recently through machine learning is that we can accurately predict customer lifetime value and, conversely, loan loss several years into the future with as little as 45 days’ worth of customer transactions. This breakthrough has enabled us to identify and engage with our most valuable customers more quickly while also predicting loan performance more accurately. This insight has had a positive impact across Enova, from marketing to risk management and finance.

Q: Sneak preview: Please tell us a take-away that you will provide during your talk at Predictive Analytics World.

A: One takeaway from our talk will be the value of looking at all the day-to-day operational decisions you make in your business and identifying those that can be automated and improved through predictive models and digital decisioning. As we’ll discuss, even areas as routine as customer payments can be improved and optimized with the right data, models, and technology.  Automated, model-driven digital decisions not only produce a better business outcome and more consistent customer experience, but they also allow businesses to be more agile and responsive to changes in the market, competition, and customer behavior.


Don’t miss Greg and Joe’s conference presentation, How Digital Decisioning Can Help Your Business Adapt Quickly to Regulatory Changes, on Wednesday, June 6, 2018 from 11:45 am to 12:05 pm at Predictive Analytics World for Financial in Las Vegas. Click here to register to attend. Use Code PATIMES for 15% off current prices (excludes workshops).

By: Eric Siegel, Founder, Predictive Analytics World

Eric Siegel is the founder of Predictive Analytics World ( — the premier machine learning conference, with cross-vendor industry events in Las Vegas, Washington DC, London, Munich, and Berlin — and the author of the award-winning book, Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie, or Die – Revised and Updated Edition, (Wiley, 2016).

Leave a Reply

Pin It on Pinterest

Share This