How Banks Can Leverage AI at the Contact Center

Banking contact centers frequency get trimmed when institutions have to cut back on expense. A veteran bank marketer and innovator says this is a misguided decision.

The primary strategic goal for most banking organizations is supposed to be meeting customer needs and wants. By and large, banks and credit unions say they differ from each other by the quality of their customer service. And yet, the contact center, tasked with providing customer service, is usually not the best funded part of the bank.

Why? Most bank executives see the contact center as a cost center and nothing else. Whenever an institution looks at efficiency the contact center, along with branches and the back office, are asked to contribute.

With the increasing interest in artificial intelligence, particularly conversational AI, the contact center is in the cross hairs for optimization and cost take outs. However, as many organizations are learning, replacing human beings with chatbots can be fraught with danger in a banking context.

Is Conversational AI Really Ready?

A salient example of the dangers of merely standing up a chatbot in lieu of a live customer service agent is the recent suit that Air Canada lost.

According to a recent story in The Guardian:

“In 2022, Jake Moffatt contacted Air Canada to determine which documents were needed to qualify for a bereavement fare, and if refunds could be granted retroactively.

“According to Moffat’s screenshot of a conversation with the chatbot, the British Columbia resident was told he could apply for the refund ‘within 90 days of the date your ticket was issued’ by completing an online form.”

The information given by the chatbot was incorrect.

Customers are to purchase discounted tickets instead of getting a refund after the flight. Moffat sued for the resulting fare difference, as Air Canada refused to honor the incorrect information given by the chatbot.

It doesn’t take an expert to imagine how something similar would play out in the banking industry. Giving incorrect information to a customer when it comes to money matters could be financially catastrophic. For this reason, many banking executives have wisely held back from jumping feet first into the conversational AI space.

According to a report issued by American Banker, “banks are proceeding with caution as a new generation of AI tools arises.” Only 6% of global/national banks (over $100 billion in assets) are moving aggressively in implementing GenAI. Virtually no banks in other categories are taking an aggressive stance. Most banks, 74%, are somewhere between learning and collecting information to doing small-scale implementations.

One of the banks that is taking a more aggressive stance is ING in the Netherlands. McKinsey reports that they have been working with ING to extend their traditional chatbot to leverage Large Language Model (LLM) capabilities. Their traditional chatbot was able to respond to 40 to 45% of the inquiries received. The remaining were sent to live agents. After 7 weeks, their generative AI powered chatbot was able to assist 20% more customers.

As we look at the growing applications of AI in the banking industry, there is a trend towards using AI, particularly LLMs, for internal optimization. For example, there is a growing number of banks putting together pilots to examine the use of Microsoft Copilot to automate and gain efficiencies. A prime candidate for these activities is, of course, the contact center.

Read more: AI Copilots Can Push Bottom-Up Innovation — But At What Cost?

What Makes the Contact Center a Key Area for Improvement?

In my career, I’ve had the opportunity to build, lead and work with contact centers within financial services. The more successful contact centers are truly customer-centric and are led and staffed by some of the most driven and knowledgeable bankers. Yet, they are usually understaffed, overworked and underpaid. They also have a dearth of modern tools.

A decade ago, I had a call center reporting up to me. It was always a Herculean challenge to get funding for new technology. Our agents had to be at their seats 15 minutes before their shift started, so they could log on to the 16 systems they had to have ready to serve our customers. A few years later I worked with a credit union where the number of systems was 18. Earlier this year, I found yet another credit union where the number of systems used to support clients is 25.

The level of inefficiency is staggering.

In passing I mentioned this inefficiency to a CTO, and his response was “Yes, they need single logins.” Clearly, we are not looking deeply into the problem. Instead of just a single login, how about reducing the number of systems?

This is one area where I see LLMs coming into play today.

Read more: From AI Hype to Real Value: Perfecting the Balancing Act

How Should Banks Deploy AI in Contact Centers Today?

Imagine an augmented contact center agent who has an AI “sidekick” that can do several things for the agent. When a customer calls, AI can authenticate the caller by using voice recognition to bypass any need for verification. As the call gets routed to the agent, AI can pop open the customer profile on the agent’s screen.

Now the agent is ready to help. As the customer describes their problem, the AI sidekick offers several solutions for the agent to choose from.

The agent can, of course, ignore the sidekick or use what is presented to them. The sidekick can also listen to the customer and help the agent with sentiment information that can alert when a customer is becoming agitated. Finally, logging the call would fall to the sidekick instead of a paper or case manager entry. To this day most contact centers I visit don’t have any real tracking of the calls and many agents still use paper to jot down notes. Most of these notes are just for the agents themselves — and are eventually shredded.

The described experience is entirely possible with existing technology. Further, the constant corrections made by the agents on the suggestions given by the AI would help in the learning process making chatbots closer to reality.

Where AI Adoption is Taking Place

These capabilities are being deployed by most of the contact center software leaders. Nextiva has a good summary of the players (admittedly, they are biased towards their own platform). However, as noted by Christina McAllister, senior analyst at Forrester in an interview with No Jitter into this use case (speaking of the in-call assistance), but I think the broader market is still a bit in flux on how they’re going to price this kind of assistance.”

The promise however is one that banks should be paying attention. The American Banker AI report indicates that 29% of bankers expect these capabilities to be implemented within the next 12 to 18 months.

Gartner has put together a report listing the 20 most promising generative AI use cases for the banking industry. In that report, the second most promising use is “Frontline AI-Copilot” while number 8 is “Banking Cost Center Assistant.”

Read more: Winning When Customers Win: Transforming Consumer Finance

What’s Coming Next for Banking and GenAI

Banking contact center managers face the dual challenge of enhancing operational efficiency and improving customer service. The strategic application of AI, particularly through Large Language Models, presents a transformative opportunity to address these challenges. By streamlining processes such as customer authentication and call logging, and providing real-time support and sentiment analysis, AI can significantly reduce inefficiencies and improve the customer experience.

Moreover, as AI systems learn and evolve, they will become increasingly adept at handling complex inquiries, further relieving the burden on human agents and allowing them to focus on higher-value tasks. Embracing AI in the contact center is not just a path to cost reduction but a vital step towards a more responsive and customer-centric banking experience. As the banking industry continues to navigate the complexities of AI adoption, those who leverage these technologies effectively will set new standards in customer service excellence.

About the author:
Alex Jimenez is a Las Vegas-based fintech consultant. He was the chief strategy officer for Finalytics.ai and Extractable. He also has held various positions at Zions Bancorp., Rockland Trust and Bank of America.

This article was originally published on . All content © 2024 by The Financial Brand and may not be reproduced by any means without permission.