Sramana Mitra: Let’s do a few interesting use cases of how your customers are using you. You may choose from whichever application area you want.
Dave Rich: I will talk about a large credit card processor. They are using our platform to do real-time scoring in a much more cost-effective way. As I mentioned earlier, the average person coming off campus is taught R. He or she is not taught SAS or SPSS. The legacy environment was SAS. If you look at the total cost of ownership, which is the scoring engine in that whole application in process is a big part of their life; it costs the service for them. The hardware and databases environment is very expensive. The programming staff to build, maintain, and support those applications is aging. SAS, being a very big part of that ecosystem without any real competitors, charges a lot for their software.
SM: But what is the customer doing with your technology? What is it that you can do that is particularly new, cutting edge and different in leveraging big data principles?
Michelle Chambers: As you know, today people are talking a lot about big data, but they don’t have a production application around big data sets and [sets in] between. Our customers tend to use larger data sets, where when they use open source R, they are bound by the in-memory communications around R. What they are looking for is to achieve additional scalability out of using Revolution R Enterprise. They are using them across the board in many different situations. We have a large agriculture chemical company that is looking to use this for their production yield crop planning. Then we have, as Dave was saying, large credit card organizations that are using it for credit risk scoring.
SM: Please take that case of the credit risk scoring, click down, and explain to me what that big data infrastructure looks like. What are the data sets, and how is the data organized differently? Because you are calling it big data as opposed to previous versions of analytics. What is cutting edge about what you are doing in those scenarios?
DR: Validus Reinsurance is one example. The challenge is to use inputs from multiple applications and third party data sources, and Validus Re’s Actuarial Group must develop custom economic capital modeling and risk measurement/management applications to support sound financial and pricing decisions. Revolution R Enterprise is used to develop open, high-performance custom correlation and simulation models that include any and all data required for the analysis. These are used for ad hoc analysis as well as systematically, driving the decision logic implemented in the company’s production system. As a result, the company has greater confidence in its risk-related and capital reserving decisions because its risk models have improved correlation capabilities and can simulate a wide variety and quantity of events.
Another example is UpStream. It is a software development (marketing attribution and optimization) company. The challenge was to economically develop a scalable, high-performing R-powered big data analytics platform on which to provide services to clients. Revolution R Enterprise leveraged RevoScaleR for big data analytics, and Hadoop for data management. Since then, the company has achieved performance and scalability required to support its growth. UpStream Software’s platform processes 50 million scores per day per client. UpStream Software saved a client $270,000 on one campaign and generated a 14% lift for another client.
Another good example is [X+1], which provides software and services for optimized digital marketing through multi-channel visitor experiences on personalized websites and real-time digital audience targeting. The challenge was [x+1]’s need for real-time analytics, automated model updates, ability to include new data types and manage fast-growing data volumes (without sacrificing performance) were not well-matched for existing closed platform analytics application. We then re-engineered the entire analytics application with Revolution R Enterprise, leveraging RevoScaleR for big data analytics, and a distributed computing platform for data management. Since then, the company has achieved performance and scalability required for growth.