What kinds of infrastructure and systems investments has the bank made to enable these kinds of capabilities?
Shah: The cloud allows us to realize the potential of data. J.P. Morgan Access Next Generation, our client Web portal, allows clients to easily access and analyze their banking information to optimize working capital. Additionally, we are leveraging cloud internally to make software and data available to developers and other end users within the bank, driving efficiencies in the delivery, maintenance and use of such services. We also are integrating data resident in various systems so that it’s available for analysis. The many-to-many data-sources-to-applications problem that is typical of complex and mature systems becomes rationalized to one-to-many, creating more efficient flows of data. We also are making investments in other capabilities important to big data, such as storage, visualization and predictive analytics.
Are you seeing any changes in the bank’s decision-making process?
Shah: The traditional approach to problem solving has been that you start with a view of the end and you tailor your analysis around that. Big data, with the ability to automatically process large amounts of information and render it in different ways, allows for the identification of opportunities in a way that doesn’t always begin with a directed question or hypothesis to be proven or disproven. We let the data “speak,” for example, revealing patterns in buying behavior or connectivity between nodes in our client network, and it leads the way. That’s the promise of big data.
[The Morality of Big Data]
What are the barriers to realizing the potential of big data, and what’s the bank doing to address these issues?
Shah: As with any analysis, there’s a junk-in, junk-out problem. The noise-to-signal ratio is pretty high in the deluge of data. So we are spending time cleaning our data so we can improve the quality of the analysis. We’re also trying to propose simple problems to big data and use agile methodologies to do it in bite-size chunks. For example, can we reduce false positives and false negatives in transaction screening? That’s a fundamental question, with an outcome, by which we can measure success or lack of success.
Another opportunity is data governance. For example, we are zealous about protecting client information, and even within our own company, it can be difficult to access information. So we’re establishing a data governance model, which is a set of principles and rules defining who owns the data, who has access to data, how it’s made available and who has the responsibility to maintain it on an ongoing basis so the clutter doesn’t become too much. This work happens up front and is done across the lines of business in partnership with the technologists. It paves the way for sustainable gains from big data efforts.
Vipul Shah has global responsibility for strategy and business development for J.P. Morgan’s Treasury Services operations. He looks for new growth opportunities, as well as ways the bank can deliver value to corporate clients from transformational technology trends such as big data.