The Next Frontier in Banking: Big Data and Artificial Intelligence

12-22-14-NarrativeSci.pngTechnological change always precedes understanding of how it will change business. The web was a research tool before we realized it would change commerce and financial transactions. Smartphones flourished as “cool devices” before drawing people away from desktops. From Facebook to wi-fi, technologies have entered the world in one form only to have us discover an unexpected use that completely changes work, life and business.

Right now, we are in the midst of a sea change with regard to three types of technologies: big data, data analytics and artificial intelligence (AI). Each holds promise for banks. The question is, how can banks use them in a way that impacts what they do?

The goal of big data is clear—to enable understanding about what’s happening with a business, with investments and with clients so that better decisions can be made. However, big data conversations tend to focus on technology infrastructures (e.g. Hadoop, MapR, cloud computing) rather than how big data can help achieve business goals. This often happens when business and technology functions work together, but the discord has been amplified in financial services because these big data projects have taken years to design and implement. The result: Lots of technology but very little satisfaction.

Banks at are at a tipping point. They must take a step back and revisit what they want technology to accomplish. I believe there are three major opportunities for big data—understanding, discovery and predictability. Each is different and needs its own focus.

We now have access to years of performance data related to sales, products, divisions and branch activities, as well as customer opinions, which improves our ability to understand, communicate and make informed decisions. AI and cognitive computing have opened up new opportunities for banks, including narratives about performance that are automatically generated by a computer. The ability to transform data into language turns what machines only previously understood into information that humans can now easily understand. In the wealth advisory space, for example, advisors can access on-demand, up-to-date tailored performance summaries for their clients, giving them the knowledge they need to make the best investment recommendations.

Another objective of big data is to find new discoveries. Using statistical techniques or machine learning, data can now be used to discover relationships between separate data points, such as customer engagement, churn, transactions, sales and success likelihood. AI then transforms the discoveries of those correlations into actual explanations of identified relationships. And in its best application, the analysis starts with a business question. For example, knowing what you want to understand (e.g. “What causes us to lose a customer?”) drives the analysis process rather than taking a random walk through the data. By letting business needs drive the process, the resulting discoveries are data relationships that can actually be used.

The third opportunity for big data is to make predictions by leveraging the results of discovery and known business rules. These predictions give financial institutions the ability to recognize and respond to situations before they become problems. This technology has helped banks identify fraudulent behavior while it is happening as well as identify what the next pattern of suspicious behavior will be.

While AI-powered technologies are newer in banking, the impact is already being seen with systems that integrate well with big data applications. For example, IBM’s Watson is already providing financial advice to returning vets. Machine learning systems are automatically learning the rules used to identify fraud and money laundering. And, narrative generation is being used to automatically provide advisors and clients with comprehensive explanations about their investments that go beyond just the numbers. For both employees and consumers, these systems are making banking more efficient by freeing people from tasks that can be handled by computers.

Bridging the gap between what machines calculate and what people can understand, big data analysis and artificial intelligence have the potential to fundamentally change our relationship with computers and data. And in doing so, computers will be able to explain everything they know to anyone who needs to know it, whenever they need to know it.

Kris Hammond