ACI Worldwide is a clear example of the power of data and machine learning in combating fraud and driving innovation in the payments industry. By harnessing real-time data and machine learning and leveraging data science expertise, ACI delivers exceptional value to its customers. I recently spoke with our latest Digital Champion, John Madden, who is chief architect for the fraud management product line at ACI. John shares his insights on being a leader, challenges in the payments industry, and how ACI addresses these challenges.
1) What does being a Digital Champion mean to you as a leader?
Being a Digital Champion is a great recognition for myself and for the team in the work that we have done to date. As a leader, I aim to foster an environment of experimentation and innovation so that we can find new and novel ways to extract information from all the data that we can gather from our solutions.
2) What challenges in the payments industry are you focused on?
The biggest challenge in the payments industry is simply the rate of change. It's an ever-evolving industry and landscape, and fraudsters are continuously evolving to exploit this changing landscape. My challenge here at ACI is to provide a platform that captures data and allows our customers to extract the required information from our solutions so that they can make real-time decisions on whether or not to accept payment transactions and orders or reject them as fraudulent.
4) What are the primary obstacles at ACI you’re encountering in harnessing the power of data?
In our 40+ years of business, we have faced several challenges. First, extracting data from various silos is a major hurdle. Second, inconsistencies in data schemas across products complicate the process. Third, once the data is cleaned, extracting meaningful insights becomes another challenge. These challenges directly impact the value we offer to customers, especially in fraud management. Having a complete picture of the payment lifecycle helps determine its validity. However, the data is scattered across multiple silos, making it difficult to combine and gain valuable insights.
5) Why did you choose to work with Apache Cassandra and DataStax?
We chose Apache Cassandra® as our data store for both our machine learning feature store and real-time order processing feature store due to its exceptional performance capabilities. ACI Fraud Management allows our customers to define an unlimited number of data aggregations to use in their risk strategies. Cassandra provides the high throughput and very low latencies required for this functionality.
DataStax has played a critical role in our journey by providing invaluable guidance, subject matter expertise, and assistance in configuring our production environments to achieve exceptional performance, high throughput, and low latency. We have also partnered with DataStax on proofs of concept that have shaped our cloud migration strategy, future technology stack, and overall strategic vision for leveraging data within our solutions. This partnership has allowed us to explore innovative approaches and align our long-term objectives with DataStax's expertise, further enhancing our capabilities and vision for the future.
6) What kind of results can you attribute to the increased use of real-time data?
The results we can attribute to using real-time data here at ACI are our best-in-class KPIs for our fraud management solutions. We have achieved remarkably low challenge rates regarding fraud detection, minimizing both false positives and false negatives. Additionally, our ability to swiftly respond to emerging fraudulent activities in the market is unparalleled.
By leveraging data effectively, we have significantly enhanced our service, positioning ourselves as a trusted partner to our customers. They can rely on us to promptly detect and combat fraud, fostering a higher level of trust and confidence in our abilities.
7) What’s next for ACI?
We're moving much more toward cloud-based solutions. We want to incorporate more data from our products and our solutions into our data warehouse. We want to offer additional data-driven services. Our patented machine learning models will continue to improve as we feed in more data and gain more insights from that data. And we would like to expand our consortium data model to include more customers and products. This will allow us to share insights across customers and payment domains without having to share the data itself.
8) What advice would you give to any other enterprise developers that are trying to work with data?
Don't try to boil the ocean. Take one piece at a time. Don't be afraid to fail and follow the data. Follow where the data insights lead you.
