Financial services firms were one of the first to be disrupted by the Internet and mobile platforms. Thus, it was one of the first to see technology-first financial services firms (fintechs) emerge to address customer pain points. High stock trading fees? eTrade solved that problem, and forced everyone to reassess the consumer stock trading revenue structures. Consumer concerns about inputting payment information into unfamiliar e-commerce sites? PayPal solved that one, motivating a revolution in embedded finance and person-to-person (P2P) payments. Given the excitement over large language models (LLMs), it’s no surprise that the sector would embrace Generative AI in fintech.
Here, we explore Generative AI in fintech starting with the ways that fintech firms can access Generative AI technology
The financial services industry has long had a “not built here” aversion to technology solutions that originate outside their organization. Fintechs—with their tech-forward bias—are no different. As a result, there are many organizations that will want to build their own Generative AI capabilities in-house.
For those that view that route as too expensive, or unfeasible given the high burden of data and technical skills, there are also scores of subscription Generative AI solutions now available. Some of them are from point solution providers who licensed GPT-3 (now 4) or another LLM like IBM’s Watson and built added features for a specific use case. Some of them, like Persado Motivation AI, are built on proprietary LLMs (Persado’s specific focus is to enable brands to deliver personalized messages that motivate customers to act at every stage of the customer experience).
In addition, many enterprise technology companies have also developed proprietary Generative AI capabilities and integrated them into their platforms. Examples include Microsoft, Adobe Firefly, and Amazon Web Services, which has a proprietary foundation model for financial institutions.
The universe of Generative AI options available for fintech firms is vast and growing. Your firm probably already has some Generative AI capabilities available in your existing environment.
With that, let’s explore some of the use cases for Generative AI in the fintech sector.
Before we dive in, we want to narrow the frame of this conversation to B2C activities. After all, the fintech sector is vast and includes a wide range of entities. Besides organizations that deliver consumer-facing services directly to end users, there are organizations that develop capabilities for financial firms to integrate into their existing offerings. The term “fintech” can even encompass traditional bank-created digital solutions, like Zelle for payments. Given that, we’re focusing here on enabling the consumer with Generative AI.
Here are five compelling ideas:
Even very financially savvy people may not be on top of their personal financial game 100% of the time. Life happens. Work happens. Kids happen. For most of us, that means it’s a win if we remember to set the automatic payments for our bills. Optimizing our investment allocations or changing our bill payment schedule to better fit with income cycles … it’s not that we don’t want those things. We either don’t pay attention to them, however, or we don’t know what the best decision would be.
That’s where GenAI could come in. Though these solutions first wowed the public for their ability to generate fluent language, they are inherently pattern recognizers. They can “read” language data and see patterns in them—including patterns in transaction flows. When combined with information on investment allocations based on a person’s goals, age, and income, they can help automate investment advice.
Nor is investing the only advice GenAI can help provide. Many people have trouble amassing “emergency” savings. That is the three months of income people are advised to have on-hand to deal with unexpected financial shocks, such as a sudden loss of income or an unexpected expense. Or, they have trouble keeping track of spending they can write off on their taxes, like donations, commuting expenses or work supplies your employer does not reimburse. AI can be fine-tuned to recognize “wasted” spending people could reallocate to savings. Or to identify tax-deductible transactions that make end-of-year tax preparation easier.
Unlike traditional big banks that appeal to the general population through a large variety of offerings, fintechs may focus on specific customer groups. For example:
By closing specific gaps in the financial services market, fintech companies appeal to a smaller subset of customers. Their digital marketing messaging needs to recognize patterns that help them reach people trying to solve a specific need. Generative AI or, better yet, Motivation AI, can help fintech brands leverage language that inspires those individuals to take action. It can also seamlessly meet your brand voice and campaign performance goals.
Leveraging AI, particularly GenAI, to identify recommendations and customer experience opportunities is just one aspect of delivering a better fintech experience. The other side of that coin is how you communicate about them. This applies to all communications. Including marketing communications (check out our CMO’s Guide to Delivering Business Value With Generative AI) to service messages. It is especially important to nail the tone and voice of a fintech offer or message when it contains personal information.
People are sensitive about their finances. While it is helpful in theory to know how to make better use of the resources you have, advice delivered in a paternalistic or aggressive way may have the opposite effect from what the fintech firm intended.
The fact that different people respond differently to language makes the challenge of constructing effective messages even more challenging. Here’s where Generative AI can help. More specifically, here is where fintechs can leverage a GenAI language platform that recognizes different consumer behaviors and fine-tunes messages to speak to an individual as if the fintech knows them personally. Persado has worked with digital-first financial institutions, including Ally Bank and SoFi, for years, helping them optimize the language they use to engage with their customers. (See this case study on Persado’s work with Ally Bank).
Different people don’t just react differently to financial services marketing messaging. Reactions to digital marketing messaging from financial services companies also change as economic conditions change. The Persado Content Intelligence team found that from Q1 2020 to Q1 2023, the top-performing emotions and bottom performing emotions across digital marketing campaigns did a complete 180-degree turn. This radical change occurred when consumers went from having less debt and more savings during the pandemic, to having less savings and more debt when inflation began to rise sharply in 2021. With the help of Persado Motivation AI, financial services companies were able to stay ahead of this change. Our customers delivered the highest-performing digital marketing messaging, even as consumer behavior toward financial services shifted.
In many organizations—even modern organizations, as many fintechs are—functional silos result in misalignment between different parts of the customer experience. You may have a CX team looking to enable customers to get the most out of the products they have. They operate through a different reporting line from the product or platform teams that build functionality into a fintech’s website or mobile app, however.
For the user experience (UX) experts on the platform side, Generative AI in fintech that has been trained to deliver optimized customer engagement can be a powerful tool. Think of its potential to fine-tune the menus, navigation, pop-ups, and other language-based features that exist within a given platform. These features should be consistent across platforms—though that doesn’t necessarily mean that they should be exactly the same.
Persado has found from working with financial institutions that the highest-performing language may differ by channel. (Hear JP Morgan Chase talk about their experience.) In fact, Urgency language—which is not usually effective in financial context—can be very effective on landing pages.
Don’t forget the potential to deliver personalized experiences through your platforms. When it comes to personalization, many fintech brands prioritize offers and overlook the everyday opportunities to personalize the customer experience. AI could equip companies to identify the types of transactions a customer is more likely to use and place them at the top of the navigation bar or in a “quick access” list on the home page. Generative AI in fintech could further deliver personalized language on web pages or app screens. It could even adjust that language based on consumer behavior in that session.
While many traditional financial institutions are catching up, one thing that separates fintechs from banks is the in-app experience. Improving the digital messaging within the app enhances the user experience and helps customers further connect with the brand.
Consumers often embrace fintechs because they offer a more convenient, more immediate, and less expensive way to fulfill everyday needs. But that value proposition comes with a major stipulation. That is, many consumers have a zero-tolerance policy when it comes to fintech service issues. After all, fintech has based its reputation on being a modern and tech-savvy provider. If it then delivers a problem-riddled customer experience, or forces customers to endure circular conversations with ineffective chatbots, the benefit-challenge tradeoff tilts in the wrong direction.
Generative AI in fintech chatbots may provide an additional tool to deliver real-time problem resolution to fintech customers. While “conversational banking” using chatbots has been part of the fintech arsenal for years, previous iterations have relied on pre-written scripts triggered by keywords. This resulted in a very frustrating experience for customers with rare but simple requests as well as for those with more complex issues. If the chatbot also made it difficult to speak to a human about the issue (as many digital-first organizations do), it became a recipe for churn.
Today’s GenAI large language models are much more fluent and more flexible in their ability to interpret the question. As a result, they are already able to handle a broader range of fintech service issues and resolve them in real time, the first time.
Again, while it would be a major improvement to answer a larger number of customer queries automatically and in real time, how you answer also matters. Brand voice, tone, and customer relationship awareness could all elevate a chatbot interaction. That makes it more human and more valuable for your brand.
It is also important to be aware of the risks and challenges with GenAI powered chatbots, including security risks. Strong security features and embedded fraud detection capabilities may be necessary to mitigate these risks.
Generative AI has transitioned from tech hype to applied technology that delivers concrete business value. The organizations experimenting and exploring ways to leverage Generative AI in fintech will be among the first to achieve results in this fast-changing space. Persado’s enterprise-grade security, compliance, and performance-driven language model is trusted by top financial services companies. To learn more about how Persado helps fintech companies leverage Generative AI to motivate customers to engage and act, reach out and schedule a demo.
The post Generative AI in Fintech: 4 Use Cases for Driving Customer Engagement appeared first on Persado.
One of the significant challenges statisticians and data scientists face is multicollinearity, particularly its most…
Artificial intelligence is not just altering the way we interact with technology; it’s reshaping the…
This post was co-written with Mickey Alon from Vidmob. Generative artificial intelligence (AI) can be…
Prompt Poet allows you to ground LLM-generated responses to a real-world data context, opening up…
NASA has three more operational Starliner missions on the books. It hasn't decided whether it…
When training a machine learning model, you may sometimes work with datasets with a large…