What Are the 3 Biggest Use Cases for GenAI in Banking?

January 2024

Fintech & Payments

Nick Maynard

VP of Fintech Market Research

Although AI and AI-based solutions had been deployed in banking for many years, the introduction of generative AI-powered capabilities have spurred a wave of increased emphasis on practicability.

Broadly speaking, generative AI use cases in banking can be divided into two categories: customer-oriented, which span from managing daily interactions with customers to offering products and/or services that are tailored to their needs, and business-oriented, which principally concerns process improvement including the enhancement of organisational productivity. Arguably, customer use cases took flight more rapidly than that of business, given the rise of digital banking and implementation of related processes in the past few years.

This article, which is based on our latest research, examines the three biggest use cases for generative AI in banking.

Customer Acquisition

Generative AI’s biggest impact across all industries is currently the most visible in sales and marketing functions, and the banking industry is no exception to this rule.

Generative AI models, especially LLMs, are utilised to analyse, summarise, and generate content from vast amounts of data, obtained from scanning external resources including social media posts, online forums, and customer reviews, facilitating identification of brand awareness, product/service sentiments, and purchasing decisions. Once patterns in data are understood and segmented, marketing teams can focus their efforts to craft acquisition campaigns for most valuable leads to maximise ROI.

Beyond optimisation of marketing efforts such as automated campaigns based on evolution and progress of customer lifecycles, generative AI can help banks and financial institutions to sustain meaningful relationships with their customers, namely, achieving relationship banking to reduce attrition. This can be accomplished by generative AI’s help in translating complex data and analytics surrounding both potential and existing customers; presenting findings in simple and actionable ways to portfolio managers.

Juniper Research believes that the emphasis on generative AI-enabled customer acquisition will be transformed into drafting of strategies to reduce cost-to-serve for both new and existing customers and lifecycle improvement in the medium term.

Customer Onboarding

Customer onboarding can be a tedious process for banks. Given the cost pressures and competition from various players in the industry, as well as regulatory obligations, both incumbent and digital banks look to improve this experience.

At a basic level, generative AI can be of assistance to customers following onboarding steps, taking the form of chat or voice bots and delivering tailored support. Taking it a level further, generative AI can be utilised in and complements identity verification checks - providing an additional step for verification and safeguarding against synthetic identity fraud and failures within KYC, KYB, and AML processes.

Generative AI can, therefore, help streamline multiple steps, with also providing additional contribution in post-onboarding process by generating personalised content to gauge customer interest. This might help increase CSAT (Customer Satisfaction) scores for banks and financial institutions, and build more engagement and loyalty.

Moreover, post-onboarding phase might entail product offerings by leveraging the learnings obtained by generative AI models through their interactions with customers during onboarding. Customers can also be encouraged to share their onboarding experiences and give NPS (Net Promoter Scores) automatically via personalised content.

Spending Insights

Like service personalisation in different stages of customer journeys, spending insights powered by generative AI can be a game changer as integrated into banks’ CRM, ERP (Enterprise Resource Planning), and other financial systems.

Generative AI can assist in breaking down both historical and current data available in banks to analyse and identify spending patterns at the most granular level. Furthermore, as LLM models can monitor mass amounts of data flows and provide immediate visibility into spending habits, this can lay the basis for other opportunities.

These include product recommendations, upselling and/or cross-selling, and inform on acquisition costs and average lifecycles of customers and products. For instance, in September 2023, Temenos launched a generative AI solution that automatically classifies and labels customers’ banking transactions that can be used as a basis for various products/services such as best product, cashflow prediction, budget advice, and customer attrition analysis among others.

Through LLM capabilities, banks are facilitated to extend credit or other related products to their customers by leveraging not only payments data, but also additional spending insights captured by Open Banking. Generative AI-driven spending insights can also lead to new product/service development to target under or overutilisation of different customer segments and increase value of existing offerings. These offerings can be enhanced with hyper-personalised marketing campaigns and assistive technologies discussed previously, streamlined, and orchestrated effectively.

Source: Generative AI in Banking Market 2024-2030

Latest Press Release: Generative AI Spending from Banking Industry to Grow by over 1,400% by 2030, as Banks Seek to Scale AI to Revolutionise Business Models

Download the Whitepaper: How Generative AI Will Transform Banking