Industries

By: Ken Pruett I thought it might be helpful to give an example of a recent performance monitoring engagement to show just how the performance monitoring process can help.  The organization to which I'm referring has been using Knowledge Based Authentication for several years. They are issuing retail credit cards for their online channel. This is an area that usually experiences a higher rate of fraud.  The Knowledge Based Authentication product is used prior to credit being issued. The performance monitoring process involved the organization providing us with a sample of approximately 120,000 records of which some were good and some were bad.  Analysis showed that they had a 25 percent referral rate -- but they were concerned about the number of frauds they were catching.  They felt that too many frauds were getting through; they believed the fraud process was probably too lenient. Based on their input, we started a detailed analytic exercise with the intention, of course, to minimize fraud losses.  Our study found that, by changing several criteria items with the set-up, the organization was able to get the tool to be more in-line with expectations.  So, by lowering the pass rate by only 9 percent they increased their fraud find rate by 27 percent.  This was much more in-line with their goals for this process. In this situation, a score was being used, in combination with the organization's customer's ability to answer questions, to determine the overall accept or refer decision.  The change to the current set-up involved requiring customers to answer at least one more question in combination with certain scores.  Although the change was minor in nature, it yielded fairly significant results. Our next step in the engagement involved looking at the questions. Analysis showed that some questions should be eliminated due to poor performance.  They were not really separating fraud; so, removing them would be beneficial to the overall process.  We also determined that some questions performed very well.  We recommended that these questions should carry a higher weight in the overall decision process.  An example would be that a customer be required to answer only two questions correct for the higher weighted questions versus three of the lesser performing questions.  The key here is to help keep pass rates up while still preventing fraud.  Striking this delicate balance is the key objective. As you can see from this example, this is an ongoing process, but the value in that process is definitely worth the time and effort.

We've recently discussed management of risk, collections strategy, credit attributes, and the like for the bank card, telco, and real estate markets. This blog will provide insights into the trends of the automotive finance market as of third quarter 2009.  In terms of credit quality, the market has been relatively steady in year-over-year comparisons.  The subprime group saw the biggest change in risk distribution from 3Q08, with a -3.74 percent shift. Overall, balances have declined to just over $673 billion (- 4 percent).  In 3Q09, banks held the largest total of outstanding automotive balances of $241 billion (with captive auto next at $203 billion).  Credit unions had the largest increase from 3Q08 (with $5 billion) and the finance/other group had the largest decrease in balances (- $23 billion). How are automotive loans performing?  Total 30- and 60-day delinquencies are still on the rise, but the rate of increase of 30-day delinquencies appears to be slowing.   New originations are dominating in the Prime plus market (66 percent), up by 10 percent.  Lending criteria has tightened and, as a result, we see scores on both new and used vehicles continue to increase.  For new buyers, over 83 percent are Prime plus.  For used buyers, over 53 percent are Prime plus.  The average credit score changed from 762 in 3Q08 to 775 in 3Q09 -- up 13 points for new vehicles.  For used vehicles in the same time period: 670 to 684, up 14 points. Lastly, let’s take a look at how financing has changed from 3Q08 to 3Q09.  The financed amounts and monthly payments have dropped year-over-year as well as the average term and average rate. Source:  State of the Automotive Finance Market, Third Quarter 2009 by Melinda Zabritski, director of Automotive Credit at Experian and Experian-Oliver Wyman Market Intelligence Reports    

Meat and potatoes Data are the meat and potatoes of fraud detection.  You can have the brightest and most capable statistical modeling team in the world.  But if they have crappy data, they will build crappy models.  Fraud prevention models, predictive scores, and decisioning strategies in general are only as good as the data upon which they are built. How do you measure data performance? If a key part of my fraud risk strategy deals with the ability to match a name with an address, for example, then I am going to be interested in overall coverage and match rate statistics.  I will want to know basic metrics like how many records I have in my database with name and address populated.  And how many addresses do I typically have for consumers?  Just one, or many?  I will want to know how often, on average, we are able to match a name with an address.  It doesn’t do much good to tell you your name and address don’t match when, in reality, they do. With any fraud product, I will definitely want to know how often we can locate the consumer in the first place.  If you send me a name, address, and social security number, what is the likelihood that I will be able to find that particular consumer in my database?  This process of finding a consumer based on certain input data (such as name and address) is called pinning.  If you have incomplete or stale data, your pin rate will undoubtedly suffer.  And my fraud tool isn’t much good if I don’t recognize many of the people you are sending me. Data need to be fresh.  Old and out-of-date information will hurt your strategies, often punishing good consumers.  Let’s say I moved one year ago, but your address data are two-years old, what are the chances that you are going to be able to match my name and address?  Stale data are yucky. Quality Data = WIN It is all too easy to focus on the more sexy aspects of fraud detection (such as predictive scoring, out of wallet questions, red flag rules, etc.) while ignoring the foundation upon which all of these strategies are built.  

In a continuation of my previous entry, I’d like to take the concept of the first-mover and specifically discuss the relevance of this to the current bank card market. Here are some statistics to set the stage: • Q2 2009 bankcard origination levels are now at 54 percent of Q2 2008 levels • In Q2 2009, bankcard originations for subprime and deep-subprime were down 63 percent from Q2 2008 • New average limits for bank cards are down 19 percent in Q2 2009 from peak in Q3 2008 • Total unused limits continued to decline in Q3 2009, decreasing by  $100 billion in Q3 2009 Clearly, the bank card market is experiencing a decline in credit supply, along with deterioration of credit performance and problematic delinquency trends, and yet in order to grow, lenders are currently determining the timing and manner in which to increase their presence in this market. In the following points, I’ll review just a few of the opportunities and risks inherent in each area that could dictate how this occurs. Lender chooses to be a first-mover: • Mining for gold – lenders currently have an opportunity to identify long-term profitable segments within larger segments of underserved consumers. Credit score trends show a number of lower-risk consumers falling to lower score tiers, and within this segment, there will be consumers who represent highly profitable relationships. Early movers have the opportunity to access these consumers with unrealized creditworthiness at their most receptive moment, and thus have the ability to achieve extraordinary profits in underserved segments. • Low acquisition costs – The lack of new credit flowing into the market would indicate a lack of competitiveness in the bank card acquisitions space. As such, a first-mover would likely incur lower acquisitions costs as consumers have fewer options and alternatives to consider. • Adverse selection - Given the high utilization rates of many consumers, lenders could face an abnormally high adverse selection issue, where a large number of the most risky consumers are likely to accept offers to access much needed credit – creating risk management issues. • Consumer loyalty – Whether through switching costs or loyalty incentives, first-movers have an opportunity to achieve retention benefits from the development of new client relationships in a vacant competitive space. Lender chooses to be a secondary or late-mover: • Reduced risk by allowing first-mover to experience growing pains before entry. The implementation of new acquisitions and risk-based pricing management techniques with new bank card legislation will not be perfected immediately. Second-movers will be able to read and react to the responses to first movers’ strategies (measuring delinquency levels in new subprime segments) and refine their pricing and policy approaches. • One of the most common first-mover advantages is the presence of switching costs by the customer. With minimal switching costs in place in the bank card industry, the ability for second-movers to deal with an incumbent is not one where switching costs are significant issues – second-movers would be able to steal market share with relative ease. • Cherry-picked opportunities – as noted above, many previously attractive consumers will have been engaged by the first-mover, challenging the second-mover to find remaining attractive segments within the market. For instance, economic deterioration has resulted in short-term joblessness for some consumers who might be strong credit risks, given the return of capacity to repay. Once these consumers are mined by the first-mover, the second-mover will likely incur greater costs to acquire these clients. Whether lenders choose to be first to market, or follow as a second-mover, there are profitable opportunities and risk management challenges associated with each strategy.  Academics and bloggers continue to debate the merits of each, (1)  but it is the ultimately lenders of today that will provide the proof.   [1] http://www.fastcompany.com/magazine/38/cdu.html  

By: Ken Pruett The use of Knowledge Based Authentication (KBA) or out of wallet questions continues to grow. For many companies, this solution is used as one of its primary means for fraud prevention.  The selection of the proper tool often involves a fairly significant due diligence process to evaluate various offerings before choosing the right partner and solution.  They just want to make sure they make the right choice. I am often surprised that a large percentage of customers just turn these tools on and never evaluate or even validate ongoing performance.  The use of performance monitoring is a way to make sure you are getting the most out of the product you are using for fraud prevention.  This exercise is really designed to take an analytical look at what you are doing today when it comes to Knowledge Based Authentication. There are a variety of benefits that most customers experience after undergoing this fraud analytics exercise.  The first is just to validate that the tool is working properly.  Some questions to ponder include: Are enough frauds being identified? Is the manual review rate in-line with what was expected?  In almost every case I have worked on as it relates to these engagements, there were areas that were not in-line with what the customer was hoping to achieve.  Many had no idea that they were not getting the expected results. Taking this one step further, changes can also be made to improve upon what is already in place.  For example, you can evaluate how well each question is performing.  The analysis can show you which questions are doing the best job at predicting fraud.  The use of better performing questions can allow you the ability to find more fraud while referring fewer applications for manual review.  This is a great way to optimize how you use the tool. In most organizations there is increased pressure to make sure that every dollar spent is bringing value to the organization.  Performance monitoring is a great way to show the value that your KBA tool is bringing to the organization.  The exercise can also be used to show how you are proactively managing your fraud prevention process.   You accomplish this by showing how well you are optimizing how you use the tool today while addressing emerging fraud trends. The key message is to continuously measure the performance of the KBA tool you are using.  An exercise like performance monitoring could provide you with great insight on a quarterly basis.  This will allow you to get the most out of your product and help you keep up with a variety of emerging fraud trends. Doing nothing is really not an option in today’s even changing environment.  

To calculate the expected business benefits of making an improvement to your decisioning strategies, you must first identify and prioritize the key metrics you are trying to positively impact.  For example, if one of your key business objectives is improved enterprise risk management, then some of the key metrics you seek to impact, in order to effectively address changes in credit score trends, could include reducing net credit losses through improved credit risk modeling and scorecard monitoring. Assessing credit risk is a key element of enterprise risk management and can addressed as part of your application risk management processes as well as other decisioning strategies that are applied at different points in the customer lifecycle. In working with our clients, Experian has identified 15 key metrics that can be positively impacted through optimizing decisions.  As you review the list of metrics below, you should identify those metrics that are most important to your organization. • Approval rates • Booking or activation rates • Revenue • Customer net present value • 30/60/90-day delinquencies • Average charge-off amount • Average recovery amount • Manual review rates • Annual application volume • Charge-offs (bad debt & fraud) • Avg. cost per dollar collected • Average amount collected • Annual recoveries • Regulatory compliance • Churn or attrition Based on Experian’s extensive experience working with clients around the world to achieve positive business results through optimizing decisions, you can expect between a 10 percent and 15 percent improvement in any of these metrics through the improved use of data, analytics and decision management software. The initial high-level business benefit calculation, therefore, is quite important and straightforward.  As an example, assume your current approval rate for vehicle loans is 65 percent, the average value of an approved application is $200 and your volume is 75,000 applications per year.  Keeping all else equal, a 10 percent improvement in your approval rates (from 65 percent to 72 percent) would generate $10.7 million in incremental business value each year ($200 x 75,000 x .65 x 1.1).  To prioritize your business improvement efforts, you’ll want to calculate expected business benefits across a number of key metrics and then focus on those that will deliver the greatest value to your organization.  

I’ve recently been hearing a lot about how bankcard lenders are reacting to changes in legislation, and recent statistics clearly show that lenders have reduced bankcard acquisitions as they retune acquisition and account management strategies for their bankcard portfolios. At this point, there appears to be a wide-scale reset of how lenders approach the market, and one of the main questions that needs to be answered pertains to market-entry timing: Should a lender be the first to re-enter the market in a significant manner, or is it better to wait, and see how things develop before executing new credit strategies? I will dedicate my next two blogs to defining these approaches and discussing them with regard to the current bankcard market. Based on common academic frameworks, today’s lenders have the option of choosing one of the following two routes: becoming a first-mover, or choosing to take the role of a secondary or late mover. Each of these roles possess certain advantages and also corresponding risks that will dictate their strategic choices: The first-mover advantage is defined as “A sometimes insurmountable advantage gained by the first significant company to move into a new market.” (1)  Although often confused with being the first-to-market, first-mover advantage is more commonly considered for firms that first substantially enter the market. The belief is that the first mover stands to gain competitive advantages through technology, economies of scale and other avenues that result from this entry strategy. In the case of the bankcard market, current trends suggest that segments of subprime and deep-subprime consumers are currently underserved, and thus I would consider the first lender to target these customers with significant resources to have ‘first-mover’ characteristics. The second-mover to a market can also have certain advantages: the second-mover can review and assess the decisions of the first-mover and develops a strategy to take advantage of opportunities not seized by the first-mover. As well, it can learn from the mistakes of the first-mover and respond, without having to incur the cost of experiential learning and possessing superior market intelligence. So, being a first-mover and second-mover can each have its advantages and pitfalls. In my next contribution, I’ll address these issues as they pertain to lenders considering their loan origination strategies for the bankcard market. (1) http://www.marketingterms.com/dictionary/first_mover_advtanage  

Conducting a validation on historical data is a good way to evaluate fraud models; however, fraud best practices dictate that a proper validation uses properly defined fraud tags. Before you can determine if a fraud model or fraud analytics tool would have helped minimize fraud losses, you need to know what you are looking for in this category.  Many organizations have difficulty differentiating credit losses from fraud losses.  Usually, fraud losses end up lumped-in with credit losses. When this happens, the analysis either has too few “known frauds” to create a business case for change, or the analysis includes a large target population of credit losses that result in poor results. By planning carefully, you can avoid this pitfall and ensure that your validation gives you the best chance to improve your business and minimize fraud losses. As a fraud best practice for validations, consider using a target population that errs on the side of including credit losses; however, be sure to include additional variables in your sample that will allow you and your fraud analytics provider to apply various segmentations to the results.  Suggested elements to include in your sample are; delinquency status, first delinquency date, date of last valid payment, date of last bad  payment and indicator of whether the account was reviewed for fraud prior to booking. Starting with a larger population, and giving yourself the flexibility to narrow the target later will help you see the full value of the solutions you evaluate and reduce the likelihood of having to do an analysis over again.  

In a previous blog, we shared ideas for expanding the “gain” to create a successful ROI to adopt new fraud best practices  to improve.  In this post, we’ll look more closely at the “cost” side of the ROI equation. The cost of the investment- The costs of fraud analytics and tools that support fraud best practices go beyond the fees charged by the solution provider.  While the marketplace is aware of these costs, they often aren’t considered by the solution providers.  Achieving consensus on an ROI to move forward with new technology requires both parties to account for these costs.  A more robust ROI should these areas: • Labor costs- If a tool increases fraud referral rates, those costs must be taken into account. • Integration costs- Many organizations have strict requirements for recovering integration costs.  This can place an additional burden on a successful ROI. • Contractual obligations- As customers look to reduce the cost of other tools, they must be mindful of any obligations to use those tools. • Opportunity costs- Organizations do need to account for the potential impact of their fraud best practices on good customers.  Barring a true champion/challenger evaluation, a good way to do this is to remain as neutral as possible with respect to the total number of fraud alerts that are generated using new fraud tools compared to the legacy process As you can see, the challenge of creating a compelling ROI can be much more complicated than the basic equation suggests.  It is critical in many industries to begin exploring ways to augment the ROI equation.  This will ensure that our industries evolve and thrive without becoming complacent or unable to stay on top of dynamic fraud trends.  

By: Wendy Greenawalt Given the current volatile market conditions and rising unemployment rates, no industry is immune from delinquent accounts. However, recent reports have shown a shift in consumer trends and attitudes related to cellular phones. For many consumers, a cell phone is an essential tool for business and personal use, and staying connected is a very high priority. Given this, many consumers pay their cellular bill before other obligations, even if facing a poor bank credit risk. Even with this trend, cellular providers are not immune from delinquent accounts and determining the right course of action to take to improve collection rates. By applying optimization, technology for account collection decisions, cellular providers can ensure that all variables are considered given the multiple contact options available. Unlike other types of services, cellular providers have numerous options available in an attempt to collect on outstanding accounts.  This, however, poses other challenges because collectors must determine the ideal method and timing to attempt to collect while retaining the consumers that will be profitable in the long term.  Optimizing decisions can consider all contact methods such as text, inbound/outbound calls, disconnect, service limitation, timing and diversion of calls.  At the same time, providers are considering constraints such as likelihood of curing, historical consumer behavior, such as credit score trends, and resource costs/limitations.  Since the cellular industry is one of the most competitive businesses, it is imperative that it takes advantage of every tool that can improve optimizing decisions to drive revenue and retention.  An optimized strategy tree can be easily implemented into current collection processes and provide significant improvement over current processes.

A recent article in the Boston Globe talked about the lack of incentive for banks to perform wide-scale real estate loan modifications due to the lack of profitability for lenders in the current government-led program structure. The article cited a recent study by the Boston Federal Reserve that noted up to 45 percent of borrowers who receive loan modifications end up in arrears again afterwards. On the other hand, around 30 percent of borrowers cured without any external support from lenders - leading them to believe that the cost and effort required modifying delinquent loans is not a profitable or not required proposition. Adding to this, one of the study’s authors was quoted as saying “a lot of people you give assistance to would default either way or won’t default either way.” The problem that lenders face is that although they have the knowledge that certain borrowers are prone to re-default, or cure without much assistance – there has been little information available to distinguish these consumers from each other.  Segmenting these customers is the key to creating a profitable process for loan modifications, since identification of the consumer in advance will allow lenders to treat each borrower in the most efficient and profitable manner. In considering possible solutions, the opportunity exists to leverage the power of credit data, and credit attributes to create models that can profile the behaviors that lenders need to isolate. Although the rapid changes in the economy have left many lenders without a precedent behavior in which to model, the recent trend of consumers that re-default is beginning to provide lenders with correlated credit attributes to include in their models. Credit attributes were used in a recent study on strategic defaulters by the Experian-Oliver Wyman Market Intelligence Reports, and these attributes created defined segments that can assist lenders with implementing profitable loan modification policies and decisioning strategies.  

By definition, “Return on Investment” is simple: (The gain from an investment - The cost of the investment) _______________________________________________ The cost of the investment With such a simple definition, why do companies that develop fraud analytics and their customers have difficulty agreeing to move forward with new fraud models and tools?   I believe the answer lies in the definition of the factors that make up the ROI equation: “The gain from an investment”- When it comes to fraud, most vendors and customers want to focus on minimizing fraud losses.  But what happens when fraud losses are not large enough to drive change? To adopt new technology it’s necessary for the industry to expand its view of the “gain.”  One way to expand the “gain” is to identify other types of savings and opportunities that aren’t currently measured as fraud losses.  These include: Cost of other tools - Data returned by fraud tools can be used to resolve Red Flag compliance discrepancies and help fraud analysts manage high-risk accounts.  By making better use of this information, downstream costs can be avoided. Other types of “bad” organizations are beginning to look at the similarities among fraud and credit losses.  Rather than identifying a fraud trend and searching for a tool to address it, some industry leaders are taking a different approach -- let the fraud tool identify the high-risk accounts, and then see what types of behavior exist in that population.  This approach helps organizations create the business case for constant improvement and also helps them validate the way in which they currently categorize losses. To increase cross sell opportunities - Focus on the “good” populations.  False positives aren’t just filtered out of the fraud review work flow, they are routed into other work flows where relationships can be expanded.    

By: Heather Grover In my previous entry, I covered how fraud prevention affected the operational side of new DDA account opening. To give a complete picture, we need to consider fraud best practices and their impact on the customer experience. As earlier mentioned, the branch continues to be a highly utilized channel and is the place for “customized service.” In addition, for retail banks that continue to be the consumer's first point of contact, fraud detection is paramount IF we should initiate a relationship with the consumer. Traditional thinking has been that DDA accounts are secured by deposits, so little risk management policy is applied. The reality is that the DDA account can be a fraud portal into the organization’s many products. Bank consolidations and lower application volumes are driving increased competition at the branch – increased demand exists to cross-sell consumers at the point of new account opening. As a result, banks are moving many fraud checks to the front end of the process: know your customer and Red Flag guideline checks are done sooner in the process in a consolidated and streamlined fashion. This is to minimize fraud losses and meet compliance in a single step, so that the process for new account holders are processed as quickly through the system as possible. Another recent trend is the streamlining of a two day batch fraud check process to provide account holders with an immediate and final decision. The casualty of a longer process could be a consumer who walks out of your branch with a checkbook in hand – only to be contacted the next day to tell that his/her account has been shut down. By addressing this process, not only will the customer experience be improved with  increased retention, but operational costs will also be reduced. Finally, relying on documentary evidence for ID verification can be viewed by some consumers as being onerous and lengthy. Use of knowledge based authentication can provide more robust authentication while giving assurance of the consumer’s identity. The key is to use a solution that can authenticate “thin file” consumers opening DDA accounts. This means your out of wallet questions need to rely on multiple data sources – not just credit. Interactive questions can give your account holders peace of mind that you are doing everything possible to protect their identity – which builds the customer relationship…and your brand.  

By: Heather Grover In past client and industry talks, I’ve discussed the increasing importance of retail branches to the growth strategy of the bank. Branches are the most utilized channel of the bank and they tend to be the primary tool for relationship expansion. Given the face-to-face nature, the branch historically has been viewed to be a relatively low-risk channel needing little (if any) identity verification – there are less uses of robust risk-based authentication or out of wallet questions. However, a now well-established fraud best practice is the process of doing proper identity verification and fraud prevention at the point of DDA account opening. In the current environment of declining credit application volumes and approval across the enterprise, there is an increased focus on organic growth through deposits.  Doing proper vetting during DDA account openings helps bring your retail process closer in line with the rest of your organization’s identity theft prevention program. It also provides assurance and confidence that the customer can now be cross-sold and up-sold to other products. A key industry challenge is that many of the current tools used in DDA are less mature than in other areas of the organization. We see few clients in retail that are using advanced fraud analytics or fraud models to minimize fraud – and even fewer clients are using them to automate manual processes - even though more than 90 percent of DDA accounts are opened manually. A relatively simple way to improve your branch operations is to streamline your existing ID verification and fraud prevention tool set: 1. Are you using separate tools to verify identity and minimize fraud? Many providers offer solutions that can do both, which can help minimize the number of steps required to process a new account; 2. Is the solution realtime? To the extent that you can provide your new account holders with an immediate and final decision, the less time and effort you’ll spend after they leave the branch finalizing the decision; 3. Does the solution provide detail data for manual review? This can help save valuable analyst time and provider costs by limiting the need to do additional searches. In my next post, we’ll discuss how fraud prevention in DDA impacts the customer experience.

The definition of account management authentication is:  Keep your customers happy, but don’t lose sight of fraud risks and effective tools to combat those risks. In my previous posting, I discussed some unique fraud risks facing institutions during the account management phase of their customer lifecycles.  As a follow up, I want to review a couple of effective tools that allow you to efficiently minimize fraud losses during post-application: Knowledge Based Authentication (KBA) — this process involves the use of challenge/response questions beyond "secret" or "traditional" internally derived questions (such as mother's maiden name or last transaction amount). This tool allows for measurably effective use of questions based on more broad-reaching data (credit and noncredit) and consistent delivery of those questions without subjective question creation and grading by call center agents. KBA questions sourced from information not easily accessible by call center agents or fraudsters provide an additional layer of security that is more impenetrable by social engineering. From a process efficiency standpoint, the use of automated KBA also can reduce online sessions for consumers, and call times as agents spend less time self-selecting questions, self-grading responses and subjectively determining next steps. Delivery of KBA questions via consumer-facing online platforms or via interactive voice response (IVR) systems can further reduce operational costs since the entire KBA process can be accommodated without call center agent involvement. Negative file and fraud database – performing checks against known fraudulent and abuse records affords institutions an opportunity to, in batch or real time, check elements such as address, phone, and SSN for prior fraudulent use or victimization.  These checks are a critical element in supplementing traditional consumer authentication processes, particularly in an account management procedure in which consumer and/or account information may have been compromised.  Transaction requests such as address or phone changes to an account are particularly low-hanging fruit as far as running negative file checks are concerned.