Data Analytics

Why Automated Predictive Analytics Is the Future of Financial Services

Financial firms know the value of data — but how do they turn prolific data sets into predictive strategy gains?

Scott Hiemstra is responsible for leading the Financial Services team for CDW. In his role as director, he is responsible for the short and long-term performance goals, strategic direction, business initiatives and talent development of the Financial Services organization.

Data forms the core of success in the financial services industry. From evolving information around market trends to client investment and customer service histories, data drives better banking strategy. As noted by a recent Accenture survey, however, while 78 percent of financial firms now use data to inform organizational operations, just 7 percent have scaled up analytics solutions to match growing demand.

To help bridge this knowledge gap, many banks have turned to predictive analytics underpinned by automation — but not always with success. Here’s a look at two emerging automation use cases and the IT infrastructure necessary to support them at scale.

How Banks Can Deploy Next Best Offer Tools

One area of expanding impact in predictive analytics automation is “next best offer” (NBO) frameworks. Already common across e-commerce platforms like Amazon, NBO tools use available information — in this case, which products consumers are looking at or have in their carts — to suggest other items that are complementary or have been purchased by other customers under similar circumstances. If a consumer is purchasing an inflatable raft, an NBO tool might suggest life jackets, paddles or air pumps.

As noted by “The Startup,” NBOs typically rely on what are known as “collaborative filtering” models that leverage massive data sets about both products and consumers. Financial firms, however, are often data rich and product lean. While banks collect massive amounts of client information to deliver better service and provide targeted recommendations, their product sets are comparatively much smaller than any e-commerce counterparts.

Here, artificial intelligence tools offer a way to bridge this gap using a “prediction model factory” approach: By leveraging extensive customer data and their historical relationships with banking experts, it’s possible to generate limited-scope product offers — such as credit card or investment options — that have a higher likelihood of conversion.

Banks Can Use Chatbots to Collect and Deliver Customer Data

There’s also a growing use case for AI-driven chatbots in banking. While this isn’t a new approach — voice recognition bots that can answer simple questions have been around for several years — more robust chatbot solutions offer a new advantage: in-depth data collection.

As noted by Forbes, although banks have substantive amounts of customer data, many don’t have the “right” data — the specific, focused information they need to make relevant and timely recommendations. This is the benefit of next-generation bots: the ability to collect data about the customer experience end-to-end, then provide that data to customer service agents on demand.

In practice, this shifts the position of chatbots along the functional financial spectrum. By moving these bots down the data stream, they can do more than act as front-end customer connections; they can help reframe consumer conversations with live human agents by providing automated, data-driven insights.

WATCH: How to use AI and automation to drive business goals.

The Infrastructure Banks Need to Take Advantage of Analytics

While both NBO solutions and predictive bot systems offer the potential to maximize revenue and boost customer engagement, banks need the right IT infrastructure to take advantage.

To optimize for automation, three considerations are critical:

Increased storage availability. As data volume and variety ramps up, banks need both onsite and cloud-based storage infrastructure capable of keeping pace. Consider the uptick of AI-driven solutions such as natural language processing (NLP) and sentiment analysis used to better understand consumer preferences and priorities. Broader data availability increases accuracy in both cases, meaning banks need agile storage infrastructure that removes potential analytic barriers.
Improved data defense. Security is also critical in a world of predictive analysis. Banks are no strangers to the need for robust security, but as the nature of collected data shifts from market generalizations to highly specific consumer data sets, there’s an emerging need for security solutions capable of defending data anywhere, anytime and on any platform. Practically speaking, improved data defense drives customer retention. Recent survey data found that consumers expect speedy response from banks with regard to any security risk, with 66 percent saying they would no longer do business if corporate breach response was slow or ineffective.
Enhanced ITSM. Banks need IT service management (ITSM) solutions that go beyond historic break-fix models to help standardize existing processes, modernize existing practices and find opportunities for innovation. Put simply, enhanced ITSM provides the responsive framework necessary to manage new predictive services and address emerging challenges.

Banks don’t want for data. What they’re often lacking, however, is a way to convert massive data sets into targeted, actionable recommendations for advisers and call center agents. The adoption of automated predictive analytics tools, such as NBO solutions and chatbot systems, paired with robust IT infrastructure can help financial firms capitalize on the true value of in-depth data collection.

This article is part of BizTech's EquITy blog series. Please join the discussion on Twitter by using the #FinanceTech hashtag.