Data Quality

From science fiction-worthy image generators to automated underwriting, artificial intelligence (AI), big data sets and advances in computing power are transforming how we play and work. While the focus in the lending space has often been on improving the AI models that analyze data, the data that feeds into the models is just as important. Enter: data-centric AI. What is a data-centric AI? Dr. Andrew Ng, a leader in the AI field, advocates for data-centric AI and is often credited with coining the term. According to Dr. Ng, data-centric AI is, ‘the discipline of systematically engineering the data used to build an AI system.’1 To break down the definition, think of AI systems as a combination of code and data. The code is the model or algorithm that analyzes data to produce a result. The data is the information you use to train the model or later feed into the model to request a result. Traditional approaches to AI focus on the code — the models. Multiple organizations download and use the same data sets to create and improve models. But today, continued focus on model development may offer a limited return in certain industries and use cases. A data-centric AI approach focuses on developing tools and practices that improve the data. You may still need to pay attention to model development but no longer treat the data as constant. Instead, you try to improve a model's performance by increasing data quality. This can be achieved in different ways, such as using more consistent labeling, removing noisy data and collecting additional data.2 Data-centric AI isn't just about improving data quality when you build a model — it's also part of the ongoing iterative process. The data-focused approach should continue during post-deployment model monitoring and maintenance. Data-centric AI in lending Organizations in multiple industries are exploring how a data-centric approach can help them improve model performance, fairness and business outcomes. For example, lenders that take a data-centric approach to underwriting may be able to expand their lending universe, drive growth and fulfill financial inclusion goals without taking on additional risk. Conventional credit scoring models have been trained on consumer credit bureau data for decades. New versions of these models might offer increased performance because they incorporate changes in the economic landscape, consumer behavior and advances in analytics. And some new models are built with a more data-centric approach that considers additional data points from the existing data sets — such as trended data — to score consumers more accurately. However, they still solely rely on credit bureau data. Explainability and transparency are essential components of responsible AI and machine learning (a type of AI) in underwriting. Organizations need to be able to explain how their models come to decisions and ensure they are behaving as expected. Model developers and lenders that use AI to build credit risk models can incorporate new high-quality data to supplement existing data sets. Alternative credit data can include information from alternative financial services, public records, consumer-permissioned data, and buy now, pay later (BNPL) data that lenders can use in compliance with the Fair Credit Reporting Act (FCRA).* The resulting AI-driven models may more accurately predict credit risk — decreasing lenders' losses. The models can also use alternative credit data to score consumers that conventional models can't score. Infographic: From initial strategy to results — with stops at verification, decisioning and approval — see how customers travel across an Automated Loan Underwriting Journey. Business benefit of using data-centric AI models Financial services organizations can benefit from using a data-centric AI approach to create models across the customer lifecycle. That may be why about 70 percent of businesses frequently discuss using advanced analytics and AI within underwriting and collections.3 Many have gone a step further and implemented AI. Underwriting is one of the main applications for machine learning models today, and lenders are using machine learning to:4 More accurately assess credit risk models. Decrease model development, deployment and recalibration timelines. Incorporate more alternative credit data into credit decisioning. AI analytics solutions may also increase customer lifetime value by helping lenders manage credit lines, increase retention, cross-sell products and improve collection efforts. Additionally, data-centric AI can assist with fraud detection and prevention. Case study: Learn how Atlas Credit, a small-dollar lender, used a machine learning model and loan automation to nearly doubled its loan approval rates while decreasing its credit risk losses. How Experian helps clients leverage data-centric AI for better business outcomes During a presentation in 2021, Dr. Ng used the 80-20 rule and cooking as an analogy to explain why the shift to data-centric AI makes sense.5 You might be able to make an okay meal with old or low-quality ingredients. However, if you source and prepare high-quality ingredients, you're already 80% of the way toward making a great meal. Your data is the primary ingredient for your model — do you want to use old and low-quality data? Experian has provided organizations with high-quality consumer and business credit solutions for decades, and our industry-leading data sources, models and analytics allow you to build models and make confident decisions. If you need a sous-chef, Experian offers services and has data professionals who can help you create AI-powered predictive analytics models using bureau data, alternative data and your in-house data. Learn more about our AI analytics solutions and how you can get started today. 1DataCentricAI. (2023). Data-Centric AI.2Exchange.scale (2021). The Data-Centric AI Approach With Andrew Ng.3Experian (2021). Global Insights Report September/October 2021.4FinRegLab (2021). The Use of Machine Learning for Credit Underwriting: Market & Data Science Context. 5YouTube (2021). A Chat with Andrew on MLOps: From Model-Centric to Data-Centric AI *Disclaimer: When we refer to “Alternative Credit Data," this refers to the use of alternative data and its appropriate use in consumer credit lending decisions, as regulated by the Fair Credit Reporting Act. Hence, the term “Expanded FCRA Data" may also apply in this instance and both can be used interchangeably.

More than half of U.S. businesses say they discuss fraud management often, making fraud detection in banking top-of-mind. Banking fraud prevention can seem daunting, but with the proper tools, banks, credit unions, fintechs, and other financial institutions can frustrate and root out fraudsters while maintaining a positive experience for good customers. What is banking fraud? Banking fraud is a type of financial crime that uses illegal means to obtain money, assets, or other property owned or held by a bank, other financial institution, or customers of the bank. This type of fraud can be difficult to detect when misclassified as credit risk or written off as a loss rather than investigated and prevented in the future. Fraud that impacts financial institutions consists of small-scale one-off events or larger efforts perpetrated by fraud rings. Not long ago, many of the techniques utilized by fraudsters required in-person or phone-based activities. Now, many of these activities are online, making it easier for fraudsters to disguise their intent and perpetrate multiple attacks at once or in sequence. Banking fraud can include: Identity theft: When a bad actor steals a consumer’s personal information and uses it to take money, open credit accounts, make purchases, and more. Check fraud: This type of fraud occurs when a fraudster writes a bad check, forges information, or steals and alters someone else’s check. Credit card fraud: A form of identity theft where a bad actor makes purchases or gets a cash advance in the name of an unsuspecting consumer. The fraudster may takeover an existing account by gaining access to account numbers online, steal a physical card, or open a new account in someone else’s name.  Phishing: These malicious efforts allow scammers to steal personal and account information through use of email, or in the case of smishing, through text messages. The fraudster often sends a link to the consumer that looks legitimate but is designed to steal login information, personally identifiable information, and more. Direct deposit account fraud: Also known as DDA fraud, criminals monetize stolen information to open new accounts and divert funds from payroll, assistance programs, and more. Unfortunately, this type of fraud doesn’t just lead to lost funds – it also exposes consumer data, impacts banks’ reputations, and has larger implications for the financial system. Today, top concerns for banks include generative AI (GenAI) fraud, peer-to-peer (P2P) payment scams, identity theft and transaction fraud. Without the proper detection and prevention techniques, it’s difficult for banks to keep fraudsters perpetrating these schemes out. What is banking fraud prevention? Detecting and preventing banking fraud consists of a set of techniques and tasks that help protect customers, assets and systems from those with malicious intent. Risk management solutions for banks identify fraudulent access attempts, suspicious transfer requests, signs of false identities, and more. The financial industry is constantly evolving, and so are fraudsters. As a result, it’s important for organizations to stay ahead of the curve by investing in new fraud prevention technologies. Depending on the size and sophistication of your institution, the tools and techniques that comprise your banking fraud prevention solutions may look different. However, every strategy should include multiple layers of friction designed to trip up fraudsters enough to abandon their efforts, and include flags for suspicious activity and other indicators that a user or transaction requires further scrutiny.   Some of the emerging trends in banking fraud prevention include: Use of artificial intelligence (AI) and machine learning (ML). While these technologies aren’t new, they are finding footing across industries as they can be used to identify patterns consistent with fraudulent activity – some of which are difficult or time-consuming to detect with traditional methods. Behavioral analytics and biometrics. By noting standard customer behaviors — e.g., which devices they use and when — and how they use those devices — looking for markers of human behavior vs. bot or fraud ring activity — organizations can flag riskier users for additional authentication and verification. Leveraging additional data sources. By looking beyond standard credit reports when opening credit accounts, organizations can better detect signs of identity theft, synthetic identities, and even potential first-party fraud.     With real-time fraud detection tools in place, financial institutions can more easily identify good consumers and allow them to complete their requests while applying the right amount and type of friction to detect and prevent fraud.   How to prevent and detect banking fraud In order to be successful in the fight against fraud and keep yourself and your customers safe, financial institutions of all sizes and types must: Balance risk mitigation with the customer experience Ensure seamless interactions across platforms for known consumers who present little to no risk Leverage proper identity resolution and verification tools Recognize good consumers and apply the proper fraud mitigation techniques to riskier scenarios With Experian’s interconnected approach to fraud detection in banking, incorporating data, analytics, fraud risk scores, device intelligence, and more, you can track and assess various activities and determine where additional authentication, friction, or human intervention is required. Learn more

Learn how high-quality data from multiple data sources can help drive business growth.

Business intelligence analytics can help financial institutions optimize their decisioning and uncover safe growth opportunities.

Explore what identity resolution for credit marketing is and how it enables lenders to create more cohesive and personalized customer interactions.

Want to retain more customers and onboard new prospects, too? A customer-driven marketing strategy might be the tool you need to boost your marketing ROI.

Putting customers at the center of your credit marketing strategy is key to achieving higher response rates and building long-term relationships. To do this, financial institutions need fresh and accurate consumer data to inform their decisions. Atlas Credit was looking to achieve higher response rates on its credit marketing campaigns by engaging consumers with timely and personalized offers. The company implemented Experian’s Ascend Marketing, a customer marketing and acquisition engine that provides marketers with accurate and comprehensive consumer credit data to build and deploy intelligent marketing campaigns. With deeper insights into their consumers, Atlas Credit created timely and customized credit offers, resulting in a 185% increase in loan originations within the first year of implementation. Additionally, the company was able to effectively manage and monitor its targeting strategies in one place, leading to improved operational efficiency and lower acquisition costs. To learn more about creating better-targeted marketing campaigns and enhancing your strategies, read the full case study. Download the case study Learn more

Financial institutions have gone through a whirlwind in the last few years, with the pandemic forcing many to undergo digital transformations. More recently, rising interest rates and economic uncertainty are leading to a pullback, highlighting the need for lenders to level up their marketing strategies to win new customers. To get started, here are a few key trends to look out for in the new year and fresh marketing ideas for lenders. Challenges and consumers expectations in 2023 It might be cliche to mention the impact that the pandemic had on digital transformations — but that doesn't make it any less true. Consumers now expect a straightforward online experience. And while they may be willing to endure a slightly more manual process for certain purchases in their life, that's not always necessary. Lenders are investing in front-end platforms and behind-the-scenes technology to offer borrowers faster and more intuitive services. For example, A McKinsey report from December 2021 highlighted the growth in nonbank mortgage lenders. It suggested nonbank lenders could hold onto and may continue taking market share as these tech-focused lenders create convenient, fast and transparent processes for borrowers.2 Marketers can take these new expectations to heart when discussing their products and services. To the extent you have one in place, highlight the digital experience that you can offer borrowers throughout the application, verifications, closing and loan servicing. You can also try to show rather than tell with interactive online content and videos. Build a data-driven mortgage lending marketing strategy The McKinsey report also highlighted a trend in major bank and nonbank lenders investing in proprietary and third-party technology and data to improve the customer experience.2 Marketers can similarly turn to a data-driven credit marketing strategy to help navigate shifting lending environments. Segment prospects with multidimensional data Successful marketers can incorporate the latest technological and multidimensional data sources to find, track and reach high-value prospects. By combining traditional credit data with marketing data and Fair Credit Report Act-compliant alternative credit data* (or expanded FCRA-regulated data), you can increase the likelihood of connecting with consumers who meet your credit criteria and will likely respond. For example, Experian's mortgage-specific In the Market Models predict a consumer's propensity to open a new mortgage within a one to four-month period based on various inputs, including trended credit data and Premier Attributes. You can use these propensity models as part of your prescreen criteria, to cross-sell current customers and to help retain customers who might be considering a new lender. But propensity models are only part of the equation, especially when you're trying to extend your marketing budget with hyper-segmented campaigns. Incorporating your internal CRM data and non-FCRA data can help you further distinguish look-alike populations and help you customize your messaging. LEARN MORE: Use this checklist to find and fix gaps in your prospecting strategy Maintain a single view of your borrowers An identity management platform can give you a single view of a consumer as they move through the customer journey. The persistent identity can also help you consistently reach consumers in a post-cookie world and contact them using their preferred channel. You can add to the persistent identity as you learn more about your prospects. However, you need to maintain data accuracy and integrity if you want to get a good ROI. Use triggers to guide your outreach You can also use data-backed credit triggers to implement your marketing plan. Experian's Prospect Triggers actively monitors a nationwide database to identify credit-active consumers who have new tradelines, inquiries or a loan nearing term. Lenders using Prospect Triggers can receive real-time or periodic updates and customize the results based on their screening strategy and criteria, such as score ranges and attributes. They can then make firm credit offers to the prospects who are most likely to respond, which can improve cross-selling opportunities along with originations. Benefit from our expertise Forward-thinking lenders should power their marketing strategies with a data-backed approach to incorporate the latest information from internal and external sources and reach the right customer at the right time and place. From list building to identity management and verification, you can turn to Experian to access the latest data and analytics tools. Learn about Experian credit prescreen and marketing solutions. Explore our credit prescreen solutions Learn about our marketing solutions 1Mortgage Bankers Association (October 2022). Mortgage Applications Decrease in Latest MBA Weekly Survey 2McKinsey & Company (2021). Five trends reshaping the US home mortgage industry

With consumers having more credit options than ever before, it’s imperative for lenders to get their message in front of ideal customers at the right time and place. But without clear insights into their interests, credit behaviors or financial capacity, you may risk extending preapproved credit offers to individuals who are unqualified or have already committed to another lender. To increase response rates and reduce wasted marketing spend, you must develop an effective customer targeting strategy. What makes an effective customer targeting strategy? A customer targeting strategy is only as good as the data that informs it. To create a strategy that’s truly effective, you’ll need data that’s relevant, regularly updated, and comprehensive. Alternative data and credit-based attributes allow you to identify financially stressed consumers by providing insight into their ability to pay, whether their debt or spending has increased, and their propensity to transfer balances and consolidate loans. With a more granular view of consumers’ credit behaviors over time, you can avoid high-risk accounts and focus only on targeting individuals that meet your credit criteria. While leveraging additional data sources can help you better identify creditworthy consumers, how can you improve the chances of them converting? At the end of the day, it’s also the consumer that’s making the decision to engage, and if you aren’t sending the right offer at the precise moment of interest, you may lose high-value prospects to competitors who will. To effectively target consumers who are most likely to respond to your credit offers, you must take a customer-centric approach by learning about where they’ve been, what their goals are, and how to best cater to their needs and interests. Some types of data that can help make your targeting strategy more customer-centric include: Demographic data like age, gender, occupation and marital status, give you an idea of who your customers are as individuals, allowing you to enhance your segmentation strategies. Lifestyle and interest data allow you to create more personalized credit offers by providing insight into your consumers’ hobbies and pastimes. Life event data, such as new homeowners or new parents, helps you connect with consumers who have experienced a major life event and may be receptive to event-based marketing campaigns during these milestones. Channel preference data enables you to reach consumers with the right message at the right time on their preferred channel. Target high-potential, high-value prospects By using an effective customer targeting strategy, you can identify and engage creditworthy consumers with the greatest propensity to accept your credit offer. To see if your current strategy has what it takes and what Experian can do to help, view this interactive checklist or visit us today. Review your customer targeting strategy Visit us

With Experian’s Prospect Triggers, this credit union was able to pinpoint consumers that met their credit criteria & were likely to respond to their offers.

Earlier this year, I explored the potential impact of the end of the current Public Health Emergency (PHE). The U.S. federal government has been operating under a PHE for COVID-19 for more than 30 consecutive months since it was initially announced in January 2020. On July 15, 2022, this PHE was renewed for a tenth time. Following this latest extension, the Centers for Medicare & Medicaid Services (CMS) has released a roadmap for the end of the COVID-19 PHE. In a related blog, they reiterate the commitment to provide a 60-day notice prior to the end of the PHE, but urge states and healthcare providers to prepare for the end “as soon as possible.” With these upcoming changes in mind, I wanted to review key areas for providers to consider as they prepare for the end of the PHE. Enrollments continue to increase, putting state budgets at risk From the start of the PHE in February 2020 through April 2022, Medicaid/Children’s Health Insurance Plan (CHIP) enrollment has increased by more than 17M people and this is affecting every state. Nearly half of all states have experienced an increase of more than 25% during this time period, with some experiencing increases of more than 40%. Given an average Medicaid cost to states of more than $8.4K per capita, that translates to an increase of billions of dollars. Once the PHE expires, states will have 12 months to redetermine eligibility for continued enrollment in the program, or risk bearing 100% of the associated cost. Preparing for the end of the PHE To avoid unnecessary expenditures and ensure that citizens are receiving access to the correct services, states will have to conduct a holistic review of their Medicaid rolls to confirm eligibility. In CMS’s guidance for states to prepare for the end of the PHE, they recommend creating an automated process to handle this unprecedented review. With the right partner, agencies can perform redeterminations of their existing registration rolls, and prepare for future services requests. The right solution can allow citizens to easily apply for benefits, triggering the automatic, real-time pull of income and employment information so that the agency can verify eligibility. Experian is a trusted government partner that is ready to assist states with preparing and automating the process for redetermination of benefits. To learn more about how Experian can assist with citizen benefit redetermination and registration efforts, visit us or request a call. Learn more

People-based marketing connects businesses with real people, helping them understand who their customers are and how to engage them in more meaningful ways.

Discover the dos and don’ts when it comes to implementing a successful strategy for income and employment verification for mortgage. Read more!

As card issuers go head-to-head in the battle to reach and connect with new consumers, they must implement more inclusive lending strategies.

Many financial institutions have made inclusion a strategic priority to expand their reach and help more U.S. consumers access affordable financial services. To drive deeper understanding, Experian commissioned Forrester to do new research to identify key focal points for firms and how they are moving the needle. The study found that more than two-thirds of institutions had a strategy created and implemented while one-quarter reported they are already up and running with their inclusion plans.1 Tapping into the underserved The research examines the importance of engaging new audiences such as those that are new to credit, lower-income, thin file, unbanked and underbanked as well as small businesses. To tap into these areas, the study outlines the need to develop new products and services, adopt willingness to change policies and processes, and use more data to drive better decisions and reach.2 Expanded data for improved risk decisioning The research underlines the use of alternative data and emerging technologies to expand reach to new audiences and assist many who have been underserved. In fact, sixty-two percent of financial institutions surveyed reported they currently use or are planning to use expanded data to improve risk profiling and credit decisions, with focus on: Banking data Cash flow data Employment verification data Asset, investments, and wealth management data Alternative financial services data Telcom and utility data3 Join us to learn more at our free webinar “Reaching New Heights Together with Financial Inclusion” where detailed research and related tools will be shared featuring Forrester’s principal analyst on Tuesday, May 24 from 10 – 11 a.m. PT. Register here for more information. Find more financial inclusion resources at www.experian.com/inclusionforward. Register for webinar Visit us 1 Based on Forrester research 2 Ibid. 3 Ibid.