Connect with us


How Banks Can Use AI to Build on Trust to Attract and Retain Customers



How Banks Can Use AI to Build on Trust to Attract and Retain Customers

By Michael Goodman, VP, Business Insights, NTT DATA Services

There is an often-used adage in business that it is easier to retain a customer, than to win a new one. The parallel to this is that existing customers are more likely to buy a new product than new customers. As marketing and advertising budgets continue to go up, banks and other financial institutions understand the value of their existing customers. The challenge here is that the core business of most banks is focused on debt products. The further into debt the customer goes, the more unhappy they become.

In the old days, banks included another core part of their business in the name of their institutions: “Trust.” There was a time when a banker knew their customers as individuals. A customer walked into their bank branch and the banker greeted them by name and even asked about their kids.

Those days may be gone, but customers still trust their banks. They trust banks with their hard-earned money and will even trust their bank with their private data under the right circumstances. In fact, a recent study by NTT DATA shows that 65% of consumers surveyed have complete trust in how their primary banks manage their personal data.

The trust your customers place in your bank can form the foundation for long-lasting customer relationships that will protect you against competitive threats from fintech startups, big data empires and non-banks looking to branch out. But only if you know how to use technology to give them what they really want—which isn’t greater debt, but personalized guidance.

Building large numbers of trusted relationships requires a commitment using artificial intelligence to understand customers as individuals. This means migrating to a new way of thinking about customers. Historically, banks have placed the majority of their customers into one of three segments based solely on assets. The less wealthy customers go into the mass market segment. The somewhat wealthy go into the mass affluent segment. The customers with more wealth go into a wealth management segment where personalized service is the norm. Banks need to move away from this model of few segments with many customers to one that better uses data and analytics to create a model of many segments with few customers. The goal is creating a model that treats each customer as a unique segment of one.

Most banks already know they need to do this, but struggle with how.

Know your customers through data

Most banks already possess a large volume of data on each of its customers. They gather data to support the account-opening process and its underlying due diligence and then gather more data throughout the life of the relationship. This is one area where financial institutions still hold the edge in the battle for data. They have a much more complete picture of your customers’ financial data than most other firms do. According to a PwC study, 76% of consumers say that sharing data is a “necessary evil, and NTT DATA found that 35% of the customers are willing to share even more data with you than they do currently if they see value in doing so.

But to use this data to bring value to their customers, banks must lay a data foundation that will allow them to generate the insights that will support their customers hopes and dreams. They need to move past their legacy data silos and application lock-up toward an environment where data is recognized for the asset it is. They should apply standard asset management principles, which banks know something about, to maximize the return on their data.

As customers continue to interact with their financial institutions, they provide additional information. By collecting these little bits of data, they support the goal of better knowing each customer. This understanding of customers is what will change the discussion from one based on debt needs to one that includes their life goals, especially when AI capabilities are used to give them meaningful financial advice

Bring meaning to the data by turning it into insights using AI

Too often, data collected about a customer is discarded or ignored – but banks have the power to change this narrative. If a bank learns that a customer has two young children, it is reasonable to assume that when those children turn 16, they will start driving. At 18, they may want to go to college. Has the bank captured the customers’ goals? If they have, what have they done with that information over the years? If a customer is wealthy enough to qualify for a personal banking services, those services will be constantly adjusted as clients move through life. AI will allow banks to scale this advice to customers that might not qualify for personal banking services. If proper data foundation is established, which includes addressing privacy and data security, AI models can be built to provide the advice bank customers say they want.

Ideally, these models will move from the reactive models we see today to models referred to as “limited memory models.” With limited memory models, the model “remembers” each exchange with a client. Rather than simply responding to current inputs as many models today do, limited memory models will evolve to leverage the transaction and communication history to make each customer feel special and that their needs are really being heard. This is exactly what personal bankers – or at least the good ones – do.

Change the dialog

Studies have repeatedly shown that bank customers would like to see their banks intervene in purchase decisions to help them stay on plan. NTT DATA found nearly half of consumers want their banks to offer guidance on their overall spending. Additionally, 39% want their banks to prevent purchases that could derail their financial goals. Although the customers want to maintain final control over a purchase decision, they do want to see the bank saying no sometimes.

A common example they give is one where the bank uses location data to detect that a customer has walked onto a car lot. The bank looks to capitalize on the opportunity by immediately pushing the latest rates for car loans. This is actually a great example of how banks miss the opportunity to establish a better relationship with their customers. Rather than push a loan, banks could stop to ask themselves what kind of lot it is and what might the customer be doing there? If it’s a used car lot and they know that one of the customers’ children recently turned 16, perhaps the customer is looking to get a first car for their child. This is an opportunity to help them finance the deal or even begin to help the customer’s child learn to manage their credit.

On the other hand, if the customer is at a Ferrari dealership, then they might be making a poor choice if they are also planning on paying college tuition in the coming years. Rather than just pushing rates for a loan that the customer might not even qualify for, you could step in to discourage a hasty purchase and advise saving for tuition. Find out if a Ferrari is on a customer’s bucket list. If it is, how much would a savings plan to help that customer achieve their dream be worth in terms of the value of the relationship?

But what about privacy?

Amid emerging privacy regulations, how can all this data be collected, and insights developed, without risking fines or other penalties? This is where trust comes back into play. In a study conducted by the Economist Intelligence Unit (EIU), 43% of executives noted that improved fraud detection and security is the greatest benefit to using AI. If banks approach the challenge thoughtfully and with customers’ interests at the center of their strategy, customers will agree to provide data and permit it to be used. This is particularly important to gain access to the customer’s data held by other banks. The key here is to not use data for marketing, but for individual relationship management. It is critical to maintain the trust and ensure the customers understand that they are in control over their data and that they are receiving value for allowing it to be used. Anonymous data can be used to build the models, which are then tailored, and insights delivered to specific customers. If banks don’t move toward leveraging AI to drive segment-of-one type relationships, other firms will figure out how to deliver those services and the financial data and customer trust that currently differentiates banks from challengers will be eroded and lost.

Editorial & Advertiser disclosure

Call For Entries

Global Banking and Finance Review Awards Nominations 2021
2021 Awards now open. Click Here to Nominate

Newsletters with Secrets & Analysis. Subscribe Now

Newsletters with Secrets & Analysis. Subscribe Now

Newsletters with Secrets & Analysis. Subscribe Now