Apply DA Tag

The fact that the last recession started right as smartphones were introduced to the world gives some perspective into how technology has changed over the past decade. Organizations need to leverage the same technological advancements, such as artificial intelligence and machine learning, to improve their collections strategies. These advanced analytics platforms and technologies can be used to gauge customer preferences, as well as automate the collections process. When faced with higher volumes of delinquent loans, some organizations rapidly hire inexperienced staff. With new analytical advancements, organizations can reduce overhead and maintain compliance through the collections process. Additionally, advanced analytics and technology can help manage customers throughout the customer life cycle. Let’s explore further: Why use advanced analytics in collections? Collections strategies demand diverse approaches, which is where analytics-based strategies and collections models come into play. As each customer and situation differs, machine learning techniques and constraint-based optimization can open doors for your organization. By rethinking collections outreach beyond static classifications (such as the stage of account delinquency) and instead prioritizing accounts most likely to respond to each collections treatment, you can create an improved collections experience. How does collections analytics empower your customers? Customer engagement, carefully considered, perhaps comprises the most critical aspect of a collections program—especially given historical perceptions of the collections process. Experian recently analyzed the impact of traditional collections methods and found that three percent of card portfolios closed their accounts after paying their balances in full. And 75 percent of those closures occurred shortly after the account became current. Under traditional methods, a bank may collect outstanding debt but will probably miss out on long-term customer loyalty and future revenue opportunities. Only effective technology, modeling and analytics can move us from a linear collections approach towards a more customer-focused treatment while controlling costs and meeting other business objectives. Advanced analytics and machine learning represent the most important advances in collections. Furthermore, powerful digital innovations such as better criteria for customer segmentation and more effective contact strategies can transform collections operations, while improving performance and raising customer service standards at a lower cost. Empowering consumers in a digital, safe and consumer-centric environment affects the complete collections agenda—beginning with prevention and management of bad debt and extending through internal and external account resolution. When should I get started? It’s never too early to assess and modernize technology within collections—as well as customer engagement strategies—to produce an efficient, innovative game plan. Smarter decisions lead to higher recovery rates, automation and self-service tools reduce costs and a more comprehensive customer view enhances relationships. An investment today can minimize the negative impacts of the delinquency challenges posed by a potential recession. Collections transformation has already begun, with organizations assembling data and developing algorithms to improve their existing collections processes. In advance of the next recession, two options present themselves: to scramble in a reactive manner or approach collections proactively. Which do you choose? Get started

Today, Experian and Oliver Wyman announced the launch of Ascend CECL ForecasterTM, a solution built to help financial institutions of all sizes more quickly and accurately forecast lifetime credit losses. The Financial Accounting Standards Board’s current expected credit loss (CECL) model has been a hot discussion topic throughout the financial services industry - first when it was announced (and considered one of the most significant accounting changes in decades), and most recently with the FASB’s delay for implementation for smaller lenders. As the compliance deadlines approach, Experian and Oliver Wyman have joined forces to help financial institutions adhere their loan portfolios to the new guidelines. Delivered through Experian’s Ascend Technology PlatformTM, Ascend CECL Forecaster is a new user-friendly, web-based application that combines Experian’s vast loan-level data and Premier AttributesSM, third-party macroeconomic data, valuation data and Oliver Wyman’s industry-leading CECL modeling methodology to accurately calculate potential losses over the life of a loan. “Ascend CECL Forecaster is a critical capability needed urgently by all lending and financial institutions,” said Ash Gupta, a Senior Advisor to Oliver Wyman and former Chief Risk Officer for American Express, in a press release. “The collaboration between Experian and Oliver Wyman allows a frictionless synthesis of industry data, capabilities and experience to serve customers in both first and second line of defense.” The premise behind the model, which will need access to more data than that used to calculate reserves under the incurred loss model, Allowance for Loan and Lease Losses (ALLL), is for financial institutions to estimate the expected loss over the life of a loan by using historical information, current conditions and reasonable forecasts. Built using advanced machine learning and statistical techniques, the web-based application maximizes the more than 15 years of historical credit data spanning previous economic cycles to help financial institutions gauge loan portfolio performance under various scenarios. Ascend CECL Forecaster does not require additional data nor does it require a secondary integration from the financial institution and enables organizations to more quickly test their portfolios under different economic factors. Moreover, financial institutions receive guidance from industry experts to assist with implementation and strategy. Additionally, Experian and Oliver Wyman will host a webinar to help financial institutions better understand and prepare for the upcoming CECL standards. Register today! Read the Press Release Register for Webinar

Consumer behavior is constantly evolving — from the channels they prefer to the economic trends spurring varying interest and activity. It’s no surprise that businesses find it challenging to know what their customers want today or tomorrow. But knowing and understanding this information is essential to growing your bottom line. Through years of working with businesses across every vertical, we’ve found that a solid approach to growing your business revolves around your customers. The better you know your customers, the better you can achieve your goals. Seeing the future. How well can you identify and rank your current customer population? Are you leveraging that insight to acquire new customers, manage current customers and prioritize collections efforts? If so, you’re probably using custom models in your business strategy. But if your organization is like many businesses, you may use a more traditional approach. In our highly competitive market, strategy and decisions must be based on the right data and insights. No excuses. The data is there, and we can help you turn it into actionable insights. Implementing a custom model can maximize your return on investment and help you make more profitable business decisions — now and in the future. No palm reading required. Without visiting your local fortuneteller, you still can predict the future. You need a model, but not the “runway” type. What constitutes a highly predictive and effective model? Many factors. A highly predictive custom model should incorporate robust data, advanced modeling methodologies, analytical expertise and attributes. Having these foundational components is essential to knowing your customers and making confident decisions. Models aren’t one-size-fits-all. When you take an innovative approach to model development, the model is targeted to support your specific business goals while providing the documentation required for regulatory reviews. Consider these items as you develop your custom model: Data — It all starts with the right data. Combining multiple data assets — your master-file data, our credit data and any additional data sources — is key to developing a robust model development sample. In other words, a model development sample should represent your future through-the-door population. Model design — To ensure the custom model is designed to help you achieve your specific goals, you’ll want to incorporate the latest analytics and modeling methodologies. An experienced analytics team will be essential here. Segmentation — With the right model development and segmentation strategies, you can identify optimal segments that will result in a more predictive custom model. This way, each consumer is scored on a scorecard developed using a credit profile similar to theirs. Validation — To ensure the model’s predictive ability and longevity, validate each custom model on a holdout sample and compare it with other scores to ensure it accounts for the current and future (through-the-door) consumer populations, as well as policy rule and behavioral changes. Regulatory review — Don’t forget about the documentation needed for compliance. While audits are unpleasant , fines and extensive scrutiny can significantly impact your business. Take your fortunetelling to the next level. Machine learning is all the rage. This cutting-edge technology can be embedded in your predictive models to help uncover patterns in data that may not be apparent otherwise. This can be done by comparing the performance of the machine learning model with your existing models. Once you know that machine learning can add the lift you’re looking for, you can apply that methodology to develop a custom model focused on stability, cost-efficiency, transparency and predictive performance. Predicting behavior across the Customer Life Cycle. How can a custom model benefit you? From improving baseline performance and increasing profitability by approving more good accounts to uncovering opportunities within your target market, custom models can provide the confidence needed to grow your business. Which one of these models can help you achieve your business goals? When it comes to accurately predicting customer behavior, you don’t need a crystal ball. You need a well-built, highly predictive custom model. Use the data that’s available to gain insight into your customers and grow your bottom line. If you need help, we’re here. We have the data, analytics and expertise to help you get started.

Digital channels undoubtedly create convenient experiences for consumers. We have the luxury of applying for loans or creating investment accounts from the comfort of home. However, the same opportunities are available to fraudsters. Fraudsters continue to find creative and innovative ways to expose vulnerabilities across all types of businesses. They prey on inexperienced or low-bandwidth teams that have not invested in the appropriate fraud tools in the past. Despite the imminent fraud risk involved, both consumers and businesses continue to embrace digital channels. With 90 percent of consumers worldwide conducting personal banking online, how do we protect these digital platforms with finite resources? A leading digital financial services company was forced to address this question when they experienced a large-scale fraud attack. But they weren’t in this fight alone. Download the full case study to see how our risk analyst used FraudNet to prevent millions of dollars in fraudulent funding. Client: A leading digital financial services company that operates with zero in-person branches with more than 7,000 employees Challenge/Objective: In October 2018, fraudsters deployed a large-scale, scripted attack against a North American financial services company. The fraud team was extremely understaffed. The fraud team was unable to detect and respond to the attack quickly. The fraudulent account opening activities eventually blended into account takeovers. Resolution: Our risk analyst worked quickly to analyze the geolocation, velocity and device rules firing within FraudNet for Account Opening. By having these rules in place, FraudNet was able to flag and outsort thousands of suspicious applications. Despite being a small team, the fraud investigators were able to work efficiently within the FraudNet workbench and review the true, high-risk applications. Results: Thanks to our risk analyst’s quick remediation and the FraudNet proprietary device rules: 23,800 fraudulent applications were outsorted for review. An estimated $35.7 million in fraudulent funding was prevented. However, the fight against fraud is ongoing. Our risk analyst continues to work closely with the fraud team to develop an effective strategy to prepare against future attacks.

If you’ve seen an uptick in photos of friends and celebrities looking older with wrinkles on your social media feeds, you’re not alone. A new free photo editor has taken the internet by a storm, featuring an AI-powered image-altering application that allows users to see their “future self.” All you have to do is upload a single photo (or few) from your camera roll to be enhanced. While this may seem like harmless fun, the app is now making headlines over increased privacy concerns about what occurs behind the scenes once users submit their selfies. Red flags were raised when multiple alleged negative implications were connected to the app – including the app’s ownership and the potential risk that the app downloaded a user’s entire photo album onto their database. In fact, the privacy concerns also prompted Democratic Party officials to implore federal agencies, including the FBI, “to look into the potential national security and privacy risks the phone app poses to the United States.” Since then, the app’s creators have addressed these concerns, stating most of the photo processing occurs in the cloud and most photos are deleted within 48 hours. Additionally, the only photos uploaded are ones that have been personally submitted by the user. Regardless, a database of user-submitted photos could be seen as a goldmine to fraudsters. In a time where personal and biometric data (including facial recognition) are some of the key ways to validate security, it’s important for consumers to be aware of how and where they’re sharing their data, whether it’s for an age-progression photo app, or their financial accounts. Consumers, businesses, financial institutions – everyone – should exhibit caution and take measures to ensure personal information remains secure and is not being used for nefarious reasons. While consumers may be aware that businesses are collecting data, companies should take steps to form digital trust with transparency. This could be achieved by: Educating consumers on how their data is being used Effectively communicating privacy policies and service terms more concisely Helping consumers feel in control of their information To learn more about research that indicates a shift to advanced authentication methods (including biometrics), fraud trends and how to combat them, download our e-book. Download Now

Friend or foe? Sophisticated criminals put a great deal of effort into creating convincing, verifiable personas (AKA synthetic identities). Once the fictional customer has embedded itself in your business, everything from the acquisition of financial instruments to healthcare benefits, utility services, and tax filings and refunds become vulnerable to synthetic identity fraud. Information attached to synthetic IDs can run several levels deep and be so complete that it includes public record data, credit information, documentary evidence and social media profiles that may even contain photo sets and historical details intended to deceive—all complicating your efforts to identify these fake customers before you do business with them. See real-world examples of how synthetic identity fraud is souring various markets – from auto and healthcare to financial services and public sector – in our tip sheet, Four common synthetic scenarios. Stopping synthetic ID fraud — at the door and thereafter. There are efforts underway in the market to collectively improve your ability to identify, shut down and prevent synthetic identities from entering your portfolio. This overall trend is great news for the future, but there are also near-term solutions you can apply to protect your business starting now. While it’s important to identify synthetic identities when they knock on your door, it’s just as important to conduct regular portfolio checkups to prevent negative impacts to your collections efforts. Every circumstance has its own unique parameters, but the overarching steps necessary to mitigate fraud from synthetic IDs remain the same: Identify current and near-term exposure using targeted segmentation analysis. Apply technology that alerts you when identity data doesn’t add up. Differentiate fraudulent identities from those simply based on bad data. Review front- and back-end screening procedures until they satisfy best practices. Achieve a “single view of the customer” for all account holders across access channels—online, mobile, call center and face-to-face. The right tools for the job. In addition to the steps mentioned above, stopping these fake customers from entering and then stealing from your organization isn’t easy—but with the right tools and strategies, it is possible. Here are a few of our top recommendations: Forensics Isolate and segment identities based on signals received during early account pathing, from both individuals and their device. For example, even sophisticated fraud networks can’t mimic natural per-device user interaction because these organizations work with hundreds or thousands of synthetic identities using just a few devices. It’s highly unlikely that multiple geographically separate account holders would share the same physical device. High-risk fraud scores Not all synthetic identity fraud manifests the same way. Using sophisticated logic and unique combinations of data, a high-risk fraud score looks at a consumer’s credit behavior and credit relationships over time to uncover previously undetectable risk. These scores are especially successful in detecting identities that are products of synthetic identity farms. And by targeting a specific data set and relationships, you can maintain a frictionless customer experience and reduce false positives. Analytics Use a solution that develops models of bad applicant behavior, then compares and scores your portfolio against these models. There isn’t a single rule for detecting fraudulent identities, but you can develop an informed set of rules and targeted models with the right service partner. Cross-referencing models designed to isolate high-risk identity theft cases, first-party or true-name fraud schemes, and synthetic identities can be accomplished in a decisioning strategy or via a custom model that incorporates the aggregate scores and attributes holistically. Synthetic identity detection rules These specialized rules consist of numerous conditions that evaluate a broad selection of consumer behaviors. When they occur in specific combinations, these behaviors indicate synthetic identity fraud. This broad-based approach provides a comprehensive evaluation of an identity to more effectively determine if it’s fabricated. It also helps reduce the incidence of inaccurately associating a real identity with a fictitious one, providing a better customer experience. Work streams Address synthetic identities confidently by applying analytics to work streams throughout the customer life cycle: Credit risk assessment Know Your Customer/Customer Identification Program checks Risk-based identity proofing and authentication Existing account management Manual reviews, investigations and charge-offs/collections activities Learn more about these tools and others that can help you mitigate synthetic identities in our white paper, Synthetic identities: getting real with customers. If your organization is like most, detecting SIDs hasn't been your top priority. So, there's no time to waste in preventing them from entering your portfolio. Criminals are highly motivated to innovate their approaches as rapidly as possible, and it’s important to implement a solution that addresses the continued rise of synthetic IDs from multiple engagement points. With the right set of analytics and decisioning tools, you can reduce exposure to fraud and losses stemming from synthetic identity attacks from the beginning and across the customer life cycle. We can help you detect and mitigate these fake customers before they become delinquent. Learn more

  You can do everything you can to prepare for the unexpected. But similar to how any first-time parent feels… you might need some help. Call in the grandparents! Experian has extensive expertise and has been around for a long time in the industry, but unlike your traditional grandparents, Experian continuously innovates, researches trends, and validates best practices in fraud and identity verification. That’s why we explored two prominent fraud reports, Javelin’s 2019 Identity Fraud Study: Fraudsters Seek New Targets and Victims Bear the Brunt and Experian’s 2019 Global Identity and Fraud Report — Consumer trust: Building meaningful relationships online, to help you identify and respond to new trends surrounding fraud. What we found – and what you need to know – is there are trends, technology and tactics that can help and hinder your fraud-prevention efforts. Consider the many digital channels available today. A full 91 percent of consumers transacted online in 2018. This presents a great opportunity for businesses to serve and develop relationships with customers. It also presents a great opportunity for fraudsters as well – as almost half of consumers have experienced a fraudulent online event. Since the threat of fraud is not impacting customers’ willingness to transact online, businesses are held responsible for adapting and evolving to not only protect their customers, but to secure their bottom line. This becomes increasingly important as fraudsters continue to target and expose vulnerabilities across inexperienced lines of businesses. Or, how about passwords. Research has shown that both businesses and consumers have greater confidence in biometrics, but neither is ready to stop using passwords. The continued reliance on traditional authentication methods is a delicate balance between security, trust and convenience. Passwords provide both authentication and consumer confidence in the online experience. It also adds friction to the user experience – and sometimes aggravation when passwords are forgotten. Advanced methods, like physical and behavioral biometrics and device intelligence, are gaining user confidence by both businesses and consumers. But a completely frictionless authentication experience can leave consumers doubting the safeness of their transaction. As you respond and adapt to our ever-evolving world, we encourage you to build and strengthen a trusted relationship with your customers through transparency. Consumers know that businesses are collection data about them. When a business is transparent about the use of that data, digital trust and consumer confidence soars. Through a stronger relationship, customers are more willing to accept friction and need fewer signs of security. Learn more about these and other trends, technology and tactics that can help and hinder your authentication efforts in our new E-book, Upcoming fraud trends and how to combat them.

Would you hire a new employee strictly by their resume? Surely not – there’s so much more to a candidate than what’s written on paper. With that being said, why would you determine your consumers’ creditworthiness based only on their traditional credit score? Resumes don’t always give you the full picture behind an applicant and can only tell a part of someone’s story, just as a traditional credit score can also be a limited view of your consumers. And lenders agree – findings from Experian’s 2019 State of Alternative Credit Data revealed that 65% of lenders are already leveraging information beyond the traditional credit report to make lending decisions. So in addition to the resume, hiring managers should look into a candidate’s references, which are typically used to confirm a candidate’s positive attributes and qualities. For lenders, this is alternative credit data. References are supplemental but essential to the resume, and allow you to gain new information to expand your view into a candidate – synonymous to alternative credit data’s role when it comes to lending. Lenders are tasked with evaluating their consumers to determine their stability and creditworthiness in an effort to prevent and reduce risk. While traditional credit data contains core information about a consumer’s credit data, it may not be enough for a lender to formulate a full and complete evaluation of the consumer. And for over 45 million Americans, the issue of having no credit history or a “thin” credit history is the equivalent of having a resume with little to no listed work experience. Alternative credit data helps to fill in the gaps, which has benefits for both lenders and consumers. In fact, 61% of consumers believe adding payment history would have a positive impact on their credit score, and therefore are willing to share their data with lenders. Alternative credit data is FCRA-compliant and includes information like alternative finance data, rental payments, utility payments, bank account information, consumer-permissioned data and full-file public records. Because this data shows a holistic view of the customer, it helps to determine their ability to repay debts and reveals any delinquent behaviors. These insights help lenders to expand their consumer lending universe– all while mitigating and preventing risk. The benefits can also be seen for home-based and small businesses. Fifty percent of all US small businesses are home-based, but many small business owners lack visibility due to their thin-file nature – making it extremely difficult to secure bank loans and capital to fund their businesses. And, younger generations and small business owners account for 58% of business owners who rely on short term lending. By leveraging alternative credit data, lenders can get greater insights into a small business owner’s credit profile and gauge risk. Entrepreneurs can also benefit from this information being used to build their credit profiles – making it easier for them to gain access to investment capital to fund their new ventures. Like a hiring manager, it’s important for lenders to get a comprehensive view to find the most qualified candidates. Using alternative credit data can expand your choices  – read our 2019 State of Alternative Credit Data Whitepaper to learn more and register for our upcoming webinar. Register Now

  Debt management is becoming increasingly complex. People don’t answer their phones anymore. There are many, many communication channels available (email, text, website, etc.) and just as many preferences from consumers regarding how they communicate. Prioritizing how much time and effort to spend on a debtor often requires help from advanced analytics and machine learning to optimize those strategies. Whether you are manually managing your collections strategies or are using advanced optimization to increase recovery rates, we’ve got keys to help you improve your recover rates. Watch our webinar, Keys to unlocking debt management success, to learn about: Minimizing the flow of accounts into collections and ensuring necessary information (e.g. risk, contact data) is used to determine the best course of action for accounts entering collections Recession readiness – prepare for the next recession to minimize impact Reducing costs and optimizing collections treatment strategies based on individual consumer circumstances and preferences Increasing recovery rates and improving customer experience by enabling consumers to interact with your organization in the most effective, efficient and non-threatening way possible Watch on-demand now>

Many may think of digital transformation in the financial services industry as something like emailing a PDF of a bank statement instead of printing it and sending via snail mail. After working with data, analytics, software and fraud-prevention experts, I have found that digital transformation is actually much more than PDFs. It can have a bigger and more positive influence on a business’s bottom line – especially when built on a foundation of data. Digital transformation is the new business model. And executives agree. Seventy percent of executives feel the traditional business model will disappear in the next five years due to digital transformation, according to recent Experian research. Our new e-book, Powering digital transformation: Transforming the customer experience with data, analytics and automation, says, “we live in a world of ‘evolve or fail.’ From Kodak to Blockbuster, we’ve seen businesses resist change and falter. The need to evolve is not new. What is new is the speed and depth needed to not only compete, but to survive. Digital startups are revolutionizing industries in months and years instead of decades and centuries.” So how do businesses evolve digitally? First, they must understand that this isn’t a ‘one-and-done’ event. The e-book suggests that the digital transformation life cycle is a never-ending process: Cleanse, standardize and enrich your data to create features or attributes Analyze your data to derive pertinent insights Automate your models and business practices to provide customer-centric experiences Test your techniques to find ways to improve Begin the process again Did you notice the key word or phrase in each of these steps is ‘data’ or ‘powered by data?’ Quality, reliable data is the foundation of digital transformation. In fact, almost half of CEOs surveyed said that lack of data or analytical insight is their biggest challenge to digital transformation. Our digital world needs better access to and insight from data because information derived from data, tempered with wisdom, provides the insight, speed and competitive advantage needed in our hypercompetitive environment. Data is the power behind digital transformation. Learn more about powering your digital transformation in our new e-book>

Consumer credit trends are continuously changing, making it imperative to keep up with the latest developments in originations, delinquencies on mortgages, credit cards and auto loans. By monitoring consumer behavior and market trends over time, you can predict and prepare for potential issues within each market. In this 30-minute webinar, our featured speakers, Gavin Harding, Experian Senior Business Consultant, and Alan Ikemura, Experian Data Analytics Senior Product Manager, reveal Q1 2019 market intelligence data and explore recent advances in consumer credit trends. Watch our on-demand webinar

Consumer credit trends and markets are constantly evolving, particularly when it comes to originations and delinquencies on mortgages, credit cards and auto loans. According to Experian research, over 2.7 million out of 89 million active automotive loans and leases are either 30 or 60 days delinquent. Triggers that signal a greater likelihood of consumers falling delinquent on loans, mortgages and credit card payments, include high-interest rates, a high utilization rate and recent derogatory trades. By tracking and forecasting consumer trends over time, you can more easily predict consumer behavior and better prepare for potential issues within each market. Join Gavin Harding, Experian Senior Business Consultant, and Alan Ikemura, Experian Data Analytics Senior Product Manager, during our live Quarterly Credit Trends webinar on May 30 at 2:00 p.m. ET. Our expert speakers will provide a view of the real estate market and share insights on the latest consumer credit trends. Highlights include: 2019 economic drivers Q1 2019 origination and delinquency trends Mortgage Home equity Bankcard Auto Register now

If you’re a credit risk manager or a data scientist responsible for modeling consumer credit risk at a lender, a fintech, a telecommunications company or even a utility company you’re certainly exploring how machine learning (ML) will make you even more successful with predictive analytics. You know your competition is looking beyond the algorithms that have long been used to predict consumer payment behavior: algorithms with names like regression, decision trees and cluster analysis. Perhaps you’re experimenting with or even building a few models with artificial intelligence (AI) algorithms that may be less familiar to your business: neural networks, support vector machines, gradient boosting machines or random forests. One recent survey found that 25 percent of financial services companies are ahead of the industry; they’re already implementing or scaling up adoption of advanced analytics and ML. My alma mater, the Virginia Cavaliers, recently won the 2019 NCAA national championship in nail-biting overtime. With the utmost respect to Coach Tony Bennett, this victory got me thinking more about John Wooden, perhaps the greatest college coach ever. In his book Coach Wooden and Me, Kareem Abdul-Jabbar recalled starting at UCLA in 1965 with what was probably the greatest freshman team in the history of basketball. What was their new coach’s secret as he transformed UCLA into the best college basketball program in the country? I can only imagine their surprise at the first practice when the coach told them, “Today we are going to learn how to put on our sneakers and socks correctly. … Wrinkles cause blisters. Blisters force players to sit on the sideline. And players sitting on the sideline lose games.” What’s that got to do with machine learning? Simply put, the financial services companies ready to move beyond the exploration stage with AI are those that have mastered the tasks that come before and after modeling with the new algorithms. Any ML library — whether it’s TensorFlow, PyTorch, extreme gradient boosting or your company’s in-house library — simply enables a computer to spot patterns in training data that can be generalized for new customers. To win in the ML game, the team and the process are more important than the algorithm. If you’ve assembled the wrong stakeholders, if your project is poorly defined or if you’ve got the wrong training data, you may as well be sitting on the sideline. Consider these important best practices before modeling: Careful project planning is a prerequisite — Assemble all the key project stakeholders, and insist they reach a consensus on specific and measurable project objectives. When during the project life cycle will the model be used? A wealth of new data sources are available. Which data sources and attributes are appropriate candidates for use in the modeling project? Does the final model need to be explainable, or is a black box good enough? If the model will be used to make real-time decisions, what data will be available at runtime? Good ML consultants (like those at Experian) use their experience to help their clients carefully define the model development parameters. Data collection and data preparation are incredibly important — Explore the data to determine not only how important and appropriate each candidate attribute is for your project, but also how you’ll handle missing or corrupt data during training and implementation. Carefully select the training and validation data samples and the performance definition. Any biases in the training data will be reflected in the patterns the algorithm learns and therefore in your future business decisions. When ML is used to build a credit scoring model for loan originations, a common source of bias is the difference between the application population and the population of booked accounts. ML experts from outside the credit risk industry may need to work with specialists to appreciate the variety of reject inference techniques available. Segmentation analysis — In most cases, more than one ML model needs to be built, because different segments of your population perform differently. The segmentation needs to be done in a way that makes sense — both statistically and from a business perspective. Intriguingly, some credit modeling experts have had success using an AI library to inform segmentation and then a more tried-and-true method, such as regression, to develop the actual models. During modeling: With a good plan and well-designed data sets, the modeling project has a very good chance of succeeding. But no automated tool can make the tough decisions that can make or break whether the model is suitable for use in your business — such as trade-offs between the ML model’s accuracy and its simplicity and transparency. Engaged leadership is important. After modeling: Model validation — Your project team should be sure the analysts and consultants appreciate and mitigate the risk of over fitting the model parameters to the training data set. Validate that any ML model is stable. Test it with samples from a different group of customers — preferably a different time period from which the training sample was taken. Documentation — AI models can have important impacts on people’s lives. In our industry, they determine whether someone gets a loan, a credit line increase or an unpleasant loss mitigation experience. Good model governance practice insists that a lender won’t make decisions based on an unexplained black box. In a globally transparent model, good documentation thoroughly explains the data sources and attributes and how the model considers those inputs. With a locally transparent model, you can further explain how a decision is reached for any specific individual — for example, by providing FCRA-compliant adverse action reasons. Model implementation — Plan ahead. How will your ML model be put into production? Will it be recoded into a new computer language, or can it be imported into one of your systems using a format such as the Predictive Model Markup Language (PMML)? How will you test that it works as designed? Post-implementation — Just as with an old-fashioned regression model, it’s important to monitor both the usage and the performance of the ML model. Your governance team should check periodically that the model is being used as it was intended. Audit the model periodically to know whether changing internal and external factors — which might range from a change in data definition to a new customer population to a shift in the economic environment — might impact the model’s strength and predictive power. Coach Wooden used to say, “It isn’t what you do. It’s how you do it.” Just like his players, the most successful ML practitioners understand that a process based on best practices is as important as the “game” itself.

For most businesses, building the best online experience for consumers requires a balance between security and convenience. But the challenge has always been finding a happy medium between the two – offering enough security that won’t get in the way of convenience and vice versa. In the past, it was always believed that one would always come at the expense of the other. But technology and innovation is changing how businesses approach security and is allowing them to give the maximum potential of both. Consumers want security AND convenience Consumers consider security and convenience as the foundation of their online experience. Findings from our 2019 Global Identity and Fraud Report revealed approximately 74 percent of consumers ranked security as the most important part of their online experience, followed by convenience. In other words, they expect businesses to provide them with both. We see this with how consumers are typically using the same security information each time they open a new digital account – out of convenience. But if one account is compromised, the consumer becomes vulnerable to possible fraudulent activity. With today’s technology, businesses can give consumers an easier and more secure way to access their digital accounts. Creating the optimal online experience More security usually meant creating more passwords, answering more security questions, completing CAPTCHA tests, etc. While consumers are willing to work through these friction-inducing methods to complete a transaction or access an account, it’s not always the most convenient process. Advanced data and technology has opened doors for new authentication methods, such as physical and behavioral biometrics, digital tokenization, device intelligence and machine learning, to maximize the potential for businesses to provide the best online experience possible. In fact, consumers have expressed greater confidence in businesses that implement these advanced security methods. Rates of consumer confidence in passwords was only 44 percent, compared to a 74 percent rate of consumer confidence in physical biometrics. Consumers are willing to embrace the latest security technology because it provides the security and convenience they want from businesses. While traditional forms of security were sufficient, advanced authentication methods have proven to be more reliable forms of security that consumers trust and can improve their online experience. The optimal online experience is a balance between security and convenience. Innovative technologies and data are helping businesses protect people’s identities and provide consumers with an improved online experience.  

Be warned. I’m a Philadelphia sports fan, and even after 13 months, I still relish in the only Super Bowl victory I’ve ever known as a fan. Having spent more than two decades in fraud prevention, I find that Super Bowl LII is coalescing in my mind with fraud prevention and lessons in defense more and more. Let me explain: It’s fourth-down-and-goal from the one-yard line. With less than a minute on the clock in the first half, the Eagles lead, 15 to 12. The easy option is to kick the field goal, take the three points and come back with a six-point advantage. Instead of sending out the kicking squad, the Eagles offense stays on the field to go for a touchdown. Broadcaster Cris Collingsworth memorably says, “Are they really going to go for this? You have to take the three!” On the other side are the New England Patriots, winners of two of the last three Super Bowls. Love them or hate them, the Patriots under coach Bill Belichick are more likely than any team in league history to prevent the Eagles from scoring at this moment. After the offense sets up, quarterback Nick Foles walks away from his position in the backfield to shout instructions to his offensive line. The Patriots are licking their chops. The play starts, and the ball is snapped — not to Foles as everyone expects, but to running back Corey Clement. Clement takes two steps to his left and tosses the ball the tight end Trey Burton, who’s running in the opposite direction. Meanwhile, Foles pauses as if he’s not part of the play, then trots lazily toward the end zone. Burton lobs a pass over pursuing defenders into Foles’ outstretched hands. This is the “Philly Special” — touchdown! Let me break this down: A third-string rookie running back takes the snap, makes a perfect toss — on the run — to an undrafted tight end. The tight end, who hasn’t thrown a pass in a game since college, then throws a touchdown pass to a backup quarterback who hasn’t caught a ball in any athletic event since he played basketball in high school. A play that has never been run by the Eagles, led by a coach who was criticized as the worst in pro football just a year before, is perfectly executed under the biggest spotlight against the most dominant team in NFL history. So what does this have to do with fraud? There’s currently an outbreak of breach-fueled credential stuffing. In the past couple of months, billions of usernames and passwords stolen in various high-profile data breaches have been compiled and made available to criminals in data sets described as “Collections 1 through 5.” Criminals acquire credentials in large numbers and attack websites by attempting to login with each set — effectively “stuffing” the server with login requests. Based on consumer propensity to reuse login credentials, the criminals succeed and get access to a customer account between 1 in 1,000 and 1 in 50 attempts. Using readily available tools, basic information like IP address and browser version are easy enough to alter/conceal making the attack harder to detect. Credential stuffing is like the Philly Special: Credential stuffing doesn’t require a group of elite all-stars. Like the Eagles’ players with relatively little experience executing their roles in the Philly Special, criminals with some computer skills, some initiative and the guts to try credential stuffing can score. The best-prepared defense isn’t always enough. The Patriots surely did their homework. They set up their defense to stop what they expected the Eagles to do based on extensive research. They knew the threats posed by every Eagle on the field. They knew what the Eagles’ coaches had done in similar circumstances throughout their careers. The defense wasn’t guessing. They were as prepared as they could have been. It’s the second point that worries me when I think of credential stuffing. Consumers reuse online credentials with alarming frequency, so a stolen set of credentials is likely to work across multiple organizations, possibly even yours. On top of that, traditional device recognition like cookies can’t identify and stop today’s sophisticated fraudsters. The best-prepared organizations feel great about their ability to stop the threats they’re aware of. Once they’ve seen a scheme, they make investments, improve their defenses, and position their players to recognize a risk and stop it. Sometimes past expertise won’t stop the play you can’t see coming.