Customer Retention in Finance with Rapid and Flexible AI and GenAI – Part 1: AI Customer Churn Prevention

Summary

Customer retention is a dynamic and important problem to solve in financial services. The customer acquisition cost, combined with cross-product churn and increased risk of unknown versus known customers, makes proactively predicting and preventing churn potentially even more important than increasing demand. Churn prevention is less expensive than acquiring new customers to replace the churners.

In this paper, we outline how TAZI's Customer Retention Solution works. The solution sits in TAZI's Solutions Library and uses AI and GenAI together to provide more holistic and accurate results. Business Analysts configure TAZI's Customer Retention Solution to fit their requirements, and TAZI makes it easy for them to deploy, monitor, and continuously adapt the solution as data and business environment change. Minimum data science and IT resources are needed.

We describe how continuous learning helps discover new evolving churn micro-segments. We also describe how business teams — marketing, operations, and customer outreach — can take churn prevention actions easily and quickly using TAZI's explainable AI.

This is the first white paper in the series. Subsequent papers will discuss how to include customer-360 communications to predict and prevent churn, how to reduce training-data requirements and increase transparency, how to increase both retention and profitability, and how to reduce retention marketing costs.

Introduction

We all know that the cost to retain a high-Lifetime-Value (LTV) customer is significantly less than the cost to acquire a new one. The financial institution has an established track record with the existing customer, and there are many unknowns and risks associated with newly acquired customers. Wouldn't it benefit your business if there was a system that monitored your existing book of renewals and easily identified those medium-to-high LTV customers that might churn?

Many financial institutions have dabbled in building customer-lifetime-value models (CLV), but rare are those that have been able to operationalize these models in a way that allows call center agents or branches to easily and effectively take the right action to prevent the client from leaving.

TAZI's AI system is designed to be understandable by business users, enabling them to trust AI and stay in sync with continuously changing business dynamics. TAZI operationalizes AI models that enable business users to receive alerts or lists of potential churning clients, recommend the right action through the right channel, and save your most valued clients.

When TAZI is deployed to reduce churn, retention rates increase, acquisition costs fall, and expense ratios improve. Imagine if you also used TAZI to help your marketing team find high-value prospects that look like your highest-LTV customers and convert them into new clients.

Customer Churn Problem Statement

Financial institutions need a mechanism to predict customer churn and take the right action to retain clients across various micro-segments. They need to take action in advance, before the client leaves, since it is nearly impossible to persuade a client who has already left to return. To take the appropriate business actions for each churn micro-segment, the retention messages generated by AI models for churn prediction must be understandable by the retention agent.

Customer churn risk varies in time and is driven by many parameters — total assets, changes in interest rates, competitors' pricing revisions or sales cycles, new regulations, customer service quality, digital banking app performance, regional economic conditions, and more.

Traditional models are not updated frequently. They are usually updated only after they fail. The updates require huge time and effort from data science teams. Traditional models are black boxes — the retention team receives client churn scores but doesn't understand why churn is happening, whether the models are right and trustworthy, or what the right prevention actions are.

TAZI Customer Retention Solution

TAZI uses an Automated Customer Retention Solution Workflow with four stages: view dashboards (spot churn trends), identify churner micro-segments (drill down into prospective churners), execute intervention strategy (perform outreach with the right message, channel, and time), and capture results (feedback loop continuously updates the AI models).

TAZI's continuous AI technology allows customer-retention models to update continuously so they can predict emerging churn trends. Churn behavior changes due to factors ranging from demographic and economic shifts to competition against the company's own product or marketing actions. TAZI helps automatically and continuously determine the level of contribution of each factor that drives churn behavior.

Retention-model explanations provide insights on churn within specific micro-segments. A micro-segment can be a class of customers of a certain demographic recently targeted by a competitor who dropped rates for that territory, or those customers with the highest propensity to churn due to poor service, staff inattention, or a small number of products in use.

For example, an explainable-AI interface might show that inactive customers between 39–42 years old churn if they have a single product and remain if they have more than two. Increasing the number of products through cross-selling could be a good strategy for this segment.

Choosing the Right Intervention

The appropriate actions for churn prevention depend on the micro-segment definition. Sending branch managers a monthly list of which customers to call with the right message and channel could be one action. Another might be sending outbound calls for a call center agent, or system-generated emails with tailored outreach messages based on that micro-segment's churn reason.

Competitor pricing changes may require actions such as policy rate-plan review, suggesting cross-sell for package discounts, or reviewing and updating existing pricing models. Increasing product usage might also prevent churn for certain segments. The cross-sell action can be scheduled directly for the segment through the explanation interface.

Example Implementation

With TAZI, the business analyst can drill down into a micro-segment to view specifics around the churners along with the most likely churn reason and the intervention strategy with the highest probability of success. Based on historical service patterns, the system recommends the right channel with the right offer.

For example, a customer might respond favorably to a multi-product discount, while another might stay with a fee waiver or a customized payment plan based on previous retention strategies that worked for similar customers.

Example Customer Churn Risk Report

• Jennie Hill — Savings, Checking · LTV 9.5 · Reason: Service · Offer: Multi-Product Discount · jenniehill@example.com
• Bill Hosket — Loan, Mortgage · LTV 9 · Reason: Interest Rate · Offer: Lower Interest Rate · 312-009-3904
• Kevin Rall — Savings, Investment · LTV 8.5 · Reason: Service · Offer: Personalized Service · 630-490-0094

ROI and Impact

TAZI's automation helps you identify customers ready to churn, provide the reason(s) for the churn, and recommend the right intervention to keep them on the books.

We can quantify the value of identifying and retaining more customers. For example, for a bank with $500M revenue, 100,000 customers, and an 8% churn rate (average annual revenue per customer ~$5K):

• ROI Scenario 1: Churn rate drops to 7% → 1,000 customers saved → 13% retention lift → ~$5M annual revenue saved
• ROI Scenario 2: Churn rate drops to 6% → 2,000 customers saved → 25% retention lift → ~$10M annual revenue saved

To understand and quantify the impact on your book of business, contact TAZI to estimate the value of saving your most valuable customers.

About TAZI

TAZI is a leading global AI and GenAI platform with out-of-the-box solutions for retention, demand, and fraud, headquartered in San Francisco. TAZI's patented adaptive and responsible AI technology has been included in 31 Gartner reports, including the Cool Vendor in Core AI Technologies (May 2019) and CAIDS Magic Quadrant (2022) honorable mention.

Founded in 2017, TAZI has a single mission: help financial institutions combine their experts' knowledge with AI and Generative AI to supercharge business teams across organizations, shaping their future while realizing direct benefits like cost reduction, increased efficiency, dynamic business insights, and business automation.

Through its easy-to-use, rapid and flexible AI and GenAI, TAZI supports business teams in financial services organizations to make smarter, more informed decisions. Using out-of-the-box solutions for Customer-360 Communications, Profitable Retention, and Targeted Marketing, TAZI's customers are creating tens of millions of dollars of value every year.

TAZI's Unique Value

• Business users control the design, implementation, monitoring, and updating of solutions — ensuring rapid value creation and reduced business risk.
• Automation and an easy UI/UX reduce tasks across the AI/GenAI lifecycle for business SMEs, analysts, data, and IT teams. Automation is ingrained into data controls and transformation, model creation and updates, dashboards, systems integrations, and monitoring.
• TAZI models learn continuously from both data and domain experts — imperative in dynamic, real-time environments.
• TAZI models provide explanations in the business domain's terminology for every result they produce.
• TAZI supports multiple (heterogeneous) data sources, structured and unstructured, utilizing AI and GenAI.