Technology & Analytical Services Blog

Unlocking Customer Insight Webinar July 14th!

Sterling Metz — Tue, 12 Jul 2011 09:05:00 +0000

Employment, income, wealth and many other data attributes about your customers are now available. How flexible is your connection to external data assets? How quickly can you implement new scoring and segmentation strategies into production? Hear how other leading organizations are transforming their systems to unlock customer insights faster and improve regulatory compliance. Learn the steps your organization can take to integrate more intelligence into your marketing and customer management activities.

Register now for this complimentary event, and learn how you can unlock deeper customer insight!

CBA Live 2011 Takeaways

Sterling Metz — Thu, 16 Jun 2011 10:20:00 +0000

CBA Live 2011 in Orlando showed a focus on action instead of talk. Numerous of the attended seminars held the themes less of "the sky is falling" and more "be pragmatic." CBA Live 2011 echoed some of what we heard at Best Practices in RETAIL FINANCIAL SERVICES. Better customer interaction is the key, whether it is through better segmentation in developing product, determining fee structure, more organized cross-selling in the branch and call centers, or better leveraging of new technologies for brand awareness.

Some of the highlights for us were:

Branch is still the preferred channel across all demographics -
It impresses us that branch network still remains the primary way consumers prefer interaction. What was really impressive was while they prefer their primary bank for home equity lending, rate and fee are their top decision criteria. Does this mean that Home Equity needs to be an anchor point of your customer centric strategy? Should your relationship pricing model center around losing a little in rate to gain share of wallet? Take it back to your customer strategy lab and let us know what you think.

Embrace the Consumer Finance Protection Bureau-
The keynote address served as a rallying cry from Richard Davis, CEO at US Bank. Banks can't just resort to business as usual and assume that the regulations will take care of themselves. We are seeing a shift in the power pendulum and as bankers; the sphere of influence has to change. Regulators and legislators have a bigger role in shaping the future of retail banking than in years past. Make sure you have an information network that keeps you aware of changes in the government and best practice trends in the industry.

Transparency Isn't All Bad -
BBVA Compass is welcoming the transparency that comes with new regulations. While bankers will have to fight to stay up to date, perhaps the trends around consumer sentiment will start to shift. When consumers trust that the fees they are paying are for value-add services and they know where their money is, they might start engaging more with their bank. This gets back to the point about social media. The hard currency in social media is trust. For your institution's brand to flourish over the long term it will have to be trustworthy, transparent, and engaging. Looking at the CFPB as the vehicle to change customer experience isn't just a good idea, it may become an imperative for future growth.

What were your takeaways and do any of these resonate with you?

This post was contributed by:

.

See YOU At CBA!

Sterling Metz — Fri, 10 Jun 2011 13:35:00 +0000

The Consumer Bankers Association is about to kick off their CBA LIVE 2011 – Shaping the Future of Retail Banking. With 8 customized tracks CBA LIVE is the banking event of 2011. Now is a time where executing business as usual is not enough and continuous learning is a necessity for bankers. Know, Catch, Keep will be there with the rest of the Equifax organization to participate as panelists, offer perspective on industry issues, and provide best practices. We strive to work on meeting tomorrow’s challenges and help you make better decisions. We are proud to support CBA as a premier sponsor and we look forward to meeting everyone there to discuss your acquisition, onboarding, and fraud concerns. If you can’t wait to learn more, watch the video below to learn more about what we will be discussing.

Offer Management Stages Part 2

Sterling Metz — Thu, 26 May 2011 07:49:00 +0000

Beyond the land of unqualified and risk-qualified offers lie the far more sophisticated and effective potentials of targeted and personalized offer management.

Stage Three – Targeted Offer Management

Stage Three is where risk is combined with additional data to build a more robust profile of the customer. For example, an understanding of personal income moves the conversation from a “can a consumer have this product” to “should a consumer have this product.” Not everyone in a risk profile is in the same financial situation. A perfect risk score and no free income indicates a waste of marketing spend and the potential aggravation of a happy DDA customer.

Differentiated assets and the corresponding analytics are the keys to seeing serious value in this space. Getting away from risk and standard-demographic data leads to the greatest value in data assets. Statisticians who know how to blend the assets pull the value from the numbers. Ultimately Stage Three is part of the journey, not the destination. Segmenting your customer base by best-offer is helpful, but once the initial “gain” is achieved, it is tough to get better. If you offer a product to everyone with an available income of $10,000 and a middle risk score today with a 5% acceptance, you will likely see the same results the next day. Until a policy starts understanding “why” an offer wasn’t accepted, offers are the only solution.

Solutions that provide the next best actions are what drive behavioral changes and increases in model results. Good segmentation doesn’t require channel integration. While the better personas equate to better offers, there is still the question of whether a customer is looking to expand his or her relationship. Ultimately segmenting “propensity to buy” needs a close understanding of the buyer’s behavior, something that defines Stage Four.

Stage Four – Personalized Offer Management

Stage Four is where an offer policy enters continual learning. By combining the segmentation of external data with the behavioral clues of internal behavior, retail banks can move past just making offers. Internal data allows for a personalized action-based system. If a customer uses a branch only to deposit the occasional personal check, it may not make sense to leverage these interactions for new offers. If customers have spent time looking at available cards at your site, it might be wise to try to get them on the phone or reach out via e-mail to let them know what products they might be interested in, based on their browsing history. For example, offering what others in their life stage have found useful, and what next steps might be. Stage Three leaves you driving the relationship based on corporate profit desires. Stage Four drives campaigns based on needs and wants of customers.

Banks can leverage a variety of tools to assist in making this vision a reality, but moving to Stage Four requires a “holistic view” most banks lack. Internal IT groups usually are not structured for the end-to-end accountability required. Traditional providers in the space may know the technology, have an armada of statisticians, but are too data-centric and force banks to pick two of the three. Getting to Stage Four requires all three.

Aite’s recent report on online marketing indicates retail banking isn’t really in Stage Four yet. The perpetual ad-fatigue customers are getting at every interaction doesn’t indicate a strong Stage Three presence. With tighter margins from growing fees, customer-centric banking is the buzzword. Until banks are marketing with strong behavioral triggers that get optimized by back-end algorithms on a regular basis, growth will remain elusive. Getting to this point is a strategic and technological process and Equifax recommends partnering with technology and analytic providers who can see the path and help your institution walk towards a targeted, best action approach.

To learn more, sign up for our monthly newsletter or follow us on twitter

This post was contributed by:

.

Offer Management in Four Stages

Sterling Metz — Fri, 20 May 2011 08:32:00 +0000

Bank marketing spend for credit cards is up, so why aren’t acceptance rates up as well?

The answer – for credit cards and other services – probably lies in gaining and using a better understanding of the four stages of offer management.

Stage One - Unqualified Offers

Stage One offer management, at its most basic, means welcoming all customers and pushing the same product to all of them. The same offer to every customer results in low life time value, extreme ad-fatigue, and potential attrition at the annoyance of being hassled about their money.

The regulatory climate that protects customers from unrequested credit examination prevents most completely unqualified offers of credit. While this stops most poor outbound marketing, inbound marketing doesn’t have this limit. Often bank tellers ask customers if they would like to apply for a credit card with their balance inquiry, circumventing the “firm offer of credit.” The only qualification might be that the customer isn’t paying a fee or in a “bad mood.”

This isn’t really offer management because there is no real “management.”

Stage Two – Risk Qualified Offers

Stage Two offer management is a result of the Fair Credit Reporting Act. Post FCRA all firm offers of credit are backed with a credit report where usage gets tracked by a CRA (Credit Reporting Agency). This permits some offer segmentation to make sure that a customer isn’t lured in with 1.9 percent only to get slapped with 5 percent when data is available. As far as the customer experience goes this is pretty safe. Risk-qualified offers also enable DDA cross-selling and point-of-sale prescreen. A customer requesting a DDA account is going to have to submit to a credit check. At that point a bank can automatically see what other products in the portfolio the customer also qualifies for to rapidly expand the relationship.

This still isn’t a “smart” offer. The bank likely isn’t checking other customer aspects such as how many credit instruments the customer has, if the product lines up with the customer’s common uses, or if the customer is satisfied with the instruments the customer has. At account opening we know very little about a customer’s behavior and unless you are augmenting with additional data, you can’t predict offer acceptance, just offer qualification. Even if a customer accepts this offering, there is nothing to say this is the “right” product. Fitting the customer in the wrong product creates a “poor fit”, lowering the likelihood that any other products will be accepted and decreasing the stickiness of the relationship.

Risk qualified offers do facilitate cross-channel integration. Unfortunately it is still only at a particular interaction and a credit card is the traditional cross-sell. If a history of extended offers is being kept and being leveraged as a “do not solicit until x days” list, it can help improve the banking experience.

There is a significant gap between Stage Two and Stage Three because of market conditions. As Stage Two was almost a forced strategy due to regulation, Stage Three is really the birth of a nuanced, “customer-centric” strategy. While not necessarily the largest technology shift, Stage Three represents a mentality shift for the banker and an analytics shift for the marketing organization.

In our next blog, we will address the next two stages – targeted offer management and personalized offer management.

To learn more, sign up for our monthly newsletter or follow us on twitter

This post was contributed by:

.

Credit Risk Policy Management Gets Easier

Sterling Metz — Thu, 05 May 2011 00:01:00 +0000

Making Better Risk Decisions with Multiple Data Set Modeling

As we saw in the previous blog regarding stopping fraud, the single biggest competitive advantage you can develop is the use of effective multiple-data set models. The ability to manage these multiple sets applies equally to making better risk decisions. Specifically, a little bit more accuracy in scale delivers large gains in profiling customers’ risk attributes.

Credit data is no longer all that is needed to determine risk. Credit data is only a shadow of the picture of customers. How are their utility bills paid, are they using a current address? Does this customer have delinquent utilities not yet in collections? Did this person ask for credit from you previously and not stay on top of their bill? You need to see income as well as debt. You need to be able to predict income from the job description. You need employment data as well as credit history. In short, credit history shows only the after-effects of all the other data points.

The knowledge of these data sources and blending them together is where value is really created. In fact the value created can be enormous.
With this augmenting information, your predictive success can improve drops by a few KS points. KS (or Kolomogorov – Smiroff) measurements are a statistician’s yardstick of how well a model predicts behavior and thus risk. In head-to-head analytic comparisons, the KS score is the gold standard in modeling.

A point or two improvement in KS scores can translate to millions of dollars. A single KS point can represent from a half percent to three percent of risky people. A three percent movement on a $1,000,000 household portfolio is going to have an incredible value. Thus finding additional data that move even a couple of percent of customers in the right direction is well worth the mining. Many companies try to find these points with custom models. When custom models fail, it can be a good idea to switch to a multi-data source set of attributes. Let’s see an example.

If a credit file has 10 trades that are 90 days past due, declining a customer is pretty easy to do. If the customer is not past due on either of only two trades, it might be tough to offer them a better rate. If you had additional data confirming they had a nice satellite and cell phone package, however, you might want to see if you can cross-sell a better rate.

It does take a lot of creativity to get to the right attributes from multiple data sources. Being a data company, Equifax gets to know customers really well. Long term relationships with these data sources have let us get really clever.

To learn more, sign up for our monthly newsletter or follow us on twitter

This post was contributed by:

.

Risk Management Decisions Can’t Be Made on IT Time (or Budget)

Sterling Metz — Thu, 06 Jan 2011 00:01:00 +0000

In these interesting times, not-so-lovingly referred to as the “new normal”, it’s the perfect storm for overloaded financial services’ IT departments. Budgets are constrained by recessionary economic growth and new, mandated requirements are being driven by increasing regulations.

In this environment, build or buy is a real easy choice – buying is a very sound decision.

Financial service providers traditionally have managed their own risk applications inside their IT organizations. The traditional IT model has been employed for decades – that is, purchase and manage the hardware and software to manage the business. However, our experience is that while they need the infrastructure to run their business – to manage the money of which they are a steward of – the same model may not be the most effective for either point-of-sale or account-management risk evaluation.

Vendor-managed applications exist to serve where internal resources fall short. Often business managers feel the IT organization is a bottleneck. Not for lack of capability, rather from the time lag in getting to projects. For example, one customer told us his IT department said it would only take two to three weeks to make the changes he needed, but there were six months of backlogged projects ahead of him. This is a particular problem with respect to risk decisions, where quick reaction to the changes in the economy, market, and competition is required. For example, when rule flows need to be changed and decision logic needs to be updated.

Vendors generally can solve problems in a matter of days - often with a far better, more- current product -- and bypass internal IT bottlenecks.
Given the rapid pace of change today, financial institutions are barely able to keep up with the speed and the scope of regulatory requirements coming out of local and federal governments. From a resource management perspective, this can tax a Chief Information Officer’s ability to meet the needs of all his internal customers, including changes in risk-decisioning.

Vendor-managed services can create and maintain solutions across multiple organizations to provide vital speed, lower costs and preserve capital. Companies can change policies, rules, even products, without overloading internal IT. This lets a company be more nimble and adaptable and ultimately, more successful.

Or, as one customer said when asked how he planned to keep up with the changing regulations simply said, “I don’t have to worry, I have Equifax to implement them in my time frame.”

If you want to talk to an Equifax specialist about how to be free of IT, please send us an e-mail. If you are interested in learning more about Equifax technologies and analytical services, please sign up for our monthly newsletter. It summarizes the new articles in our blog.

This article was contributed by .

Lee Grice has been product manager of Equifax's InterConnect decisioning platform since it's inception. He has been with Equifax for over 20 years. Lee graduated from Auburn with a degree in Computer Engineering.

Looking Forward By Looking Back

Sterling Metz — Thu, 30 Dec 2010 00:01:00 +0000

We hope you are having a happy and safe holiday season. As Know, Catch, Keep gears up plans for 2011, we thought we would review the posts that have stood out.

Is your risk policy getting that much needed proactive health check?
Tom talked about what keeping your model up to date with advanced analytics means for your business.

Top 5 Fraud Questions for Risk Officers
We reviewed the 5 burning questions every policy needs to answer.

KBA and MFA – The Alphabet Soup of Authentication
Following up on Equifax's big Identity fraud software acquisition, we broke down what all those letters mean.

If you are interested in learning more about Equifax technologies and advanced analytical services, please sign up for our monthly newsletter. It summarizes the new articles in our blog.

Is your risk policy getting that much needed proactive health check?

Sterling Metz — Thu, 02 Dec 2010 00:01:00 +0000

Given the enormous upheavals in all aspects of financial commerce over the last couple of years, anyone using a risk analysis model that is not frequently tested is courting trouble…if not disaster.

Sure, your model is working well, or well enough, you say. And all those upheavals mean cash is scarce and the old saw of “if it ain’t broke, don’t fix it,” holds sway. Why should you spend scarce dollars validating, and possibly refining a perfectly predictive model?

It can be a tough sell, we know. But, reality is that behaviors are constantly shifting. People change behaviors in response to changes in circumstances and environments. You have to move with them to give your company the best risk management guidance.

For example, in the past, people always paid the mortgage first. That was sacrosanct for generations. What do they pay first now? Bankcards, their lifeblood in hard times. And further heresy, the stigma of walking away from underwater homes is gone. And so are an ever-growing number of ex-homeowners. A total erosion of an almost-sacred American value in literally the blink of an eye.

Changes at the macro level like these are relatively easy to see now, but would it not have been great to have had a metaphorical tripwire out there a year ago? And who knows what lesser, more subtle changes are occurring? They are occurring and they are affecting your risk analysis model.

You need to have a way to find these micro changes before they bite you.
Prudent risk management requires frequent validation of whether your policies and processes are still working as intended. Leveraging advanced predictive analytics is crucial. For example, suppose you have a policy of sending a letter to a group of consumers reminding them to pay their mortgage. If that process traditionally results in x% success, would it not be prudent – in light of changing behaviors – to validate whether that policy is still effective?

To us, tracking the effectiveness of a policy over time would seem in order. Ideally, these efforts would lead to refinements to increase predictive capabilities. Clearly, not validating and refining your models risks significant negative ramifications. We believe validating, tracking, and refining are the most essential parts of any advanced analytical program with quick refinements being the most important.

Fortunately, there are ways to go about it that won’t break your budget.
For example, we can help you test ideas against small, sub-population samples of your model; the equivalent of a medical trial where some get the treatment and some get a placebo. This actually is a fine way to approach this dilemma of refinement. And it lets you keep your control group intact while trying out new strategies at minimal cost.

When you hit a winner, flash updates can boost your model’s effectiveness almost instantly. No reason to reinvent the wheel.

A well-monitored portfolio that reflects current business intelligence delivers a better result.

If you want to talk to an Equifax specialist about validating your models, please send us an e-mail. If you are interested in learning more about Equifax technologies and analytical services, please sign up for our monthly newsletter. It summarizes the new articles in our blog.

This article was contributed by Tom Aliff.

Tom Aliff is the Sr. Director of Analytics at Equifax. Tom has been working in statistical modeling for 7 years with experience with financial service providers. Tom attained his Bachelors and Masters in statistics from Purdue University.

Why Real-time Prescreen-of-One Should Get a Bigger Share of Your Marketing Budget

Sterling Metz — Tue, 17 Aug 2010 11:20:00 +0000

Faced with lower response to direct mail credit offers as well as economic pressures to reduce marketing costs, lenders are looking for practical strategies to meet growth objectives. Many lenders are adopting more sophisticated direct mail prescreen practices, such as lifestyle analytics, in an attempt to reach the right prospects. While effective, these “off-line” prescreens, as they are known, are only one part of the solution. The bigger opportunity for lenders is to augment the off-line prescreens with real-time — whenever the opportunity arises.

While the terms “prescreen” and “cross-sell” are often used interchangeably, there are differences between the two. Cross-selling generally occurs when a consumer has made an application for a specific product. As part of the application, the lender pulls a credit report to determine whether or not the individual is eligible for that specific product. The lender can then evaluate other products and services for cross-selling.

With an online prescreen solution, lenders can make pre-approved credit offers within seconds. Prescreen-of-one, on the other hand, leverages the consumer’s presence at the point-of-sale even though they haven’t applied for any specific product. During a real-time prescreen transaction, a consumer’s creditworthiness is evaluated in seconds. When a consumer passes the prescreen criteria, federal law dictates that a firm offer of credit must be made.

Real-time or “online” prescreen is one of the most cost-effective and flexible ways to quickly increase wallet share. However, direct mail is still a substantial source of revenue for lenders big and small. The best approach is to deploy a balanced blend of both strategies.

For more information, read our banking pre-screen-of-one white paper

Increased on-line threats. New FFIEC guidance. Expensive security upgrades? Not necessarily ...

Karthik Mani — Tue, 05 Jul 2011 07:40:00 +0000

According to new research seen in BAI banking strategies June 2011, between 2003 and 2010, the average U.S. bank branch experienced a roughly one-third decline in the number of daily sales transactions generated by each non-teller staff member. That shows a significant number of transactions moving on-line. Also, according to that same research, roughly 80% of the daily activity in a typical teller line today is manual deposit-taking and check-cashing which could easily be migrated on-line. This points to further movement of transactions on-line increasing the number of users and number of transactions on-line

Federal Financial Institutions Examination Council (FFIEC) has released the long-awaited supplement to its "Authentication in an Internet Banking Environment guidance", which was first issued by the FFIEC in October 2005. The guidance acknowledges that since 2005, there have been significant changes in the threat landscape and that malware can compromise some of the most robust online authentication techniques, including some forms of multi-factor authentication. The guidance also challenges the effectiveness of certain common authentication techniques including simple device identification and basic challenge questions.

On the other hand, most financial institutions are using the same methods that the FFIEC criticizes above as insufficient. According to Gartner, 73% of banks use Flash objects or cookies to identify devices, and 89% use challenge questions to verify the identity of the individual. For example, according to the FFIEC and our experts, experience has shown cookies may be copied and moved to a fraudster’s PC, allowing the fraudster to impersonate the legitimate customer.

Given these two trends, and constrained budgets, what can be done? You need to tighten the initial identification of a person (Identity Proofing). You need to provide strong authentication with layered security for on-going authentication. How much is this going to cost? Not as much as you think when you consider the Identity as a Service (IDaaS) choices. In addition sharing trusted identities across enterprises improves access for users and can increase trust through better identity proofing. Register for July 20 "On Demand Identity" webinar to learn more.

FFIEC releases long awaited guidance on authentication

Karthik Mani — Tue, 28 Jun 2011 16:31:00 +0000

Federal Financial Institutions Examination Council(FFIEC) has released the long-awaited supplement to its "Authentication in an Internet Banking Environment guidance", which was first issued by the FFIEC in October 2005. According to the FFIEC, the purpose of the Supplement to the 2005 Guidance (Supplement) is to reinforce the Guidance’s risk management framework and update the Agencies’ expectations regarding customer authentication, layered security, or other controls in the increasingly hostile online environment.

Given the breaches at RSA which affects many of the financial institutions that use hard tokens from RSA, and the increased threats through other vectors, this clarity and certainty is sure to be a welcome addition for banks and other financial institutions. The guidance acknowledges that since 2005, there have been significant changes in the threat landscape and that malware can compromise some of the most robust online authentication techniques, including some forms of multi-factor authentication. The guidance also challenges the effectiveness of certain common authentication techniques including simple device identification and basic challenge questions.

According to this guidance FFIEC believes that institutions should no longer rely on one form of customer authentication. They say that one dimensional customer authentication program is simply not robust enough to provide the level of security that customers expect and that protects institutions from financial and reputation risk. Equifax agrees with this and has developed solutions that utilizes multiple authentication vectors which are used selectively based on the risk. Equifax provides the layered security solution which has been implemented at banks, government and healthcare organizations. The solution meets the needs of financial institutions with respect to the new FFIEC guidance. We can help you in the journey to quickly implement a solution that meets the guidance. To learn more contact us at Technology@equifax.com.

Also sign up for our monthly newsletter or follow us on twitter for additional updates.

Data Breaches and the Banking Industry

Sterling Metz — Fri, 17 Jun 2011 08:17:00 +0000

Over the last six years, there have been 288 publicly disclosed breaches at financial services companies that exposed at least 83 million customer records, according to the Identity Theft Resource Center. The RSA Breach in March has elevated the interest of this issue but getting useful information is filtering the signal from the noise. The Citi Breach is something all bankers should keep a close eye on. We are going to do our best to keep updating this post with relevant stories.

Token replacement could take 6-8 months and isn't free for banks.

Token replacement looks like it could cost the banking industry between 50 and 100 million dollars

Citi's breach looks to now have effected 360,000 instead of the 200,000 initially estimated.

Citi's breach looks to have it's origins in eastern Europe and the hackers weren't impressed with Citi's defenses.

Deloitte reports that of all nations, the United States has the most financial institutions that were still "catching up" on security.

Small banks are stuck between a rock and a hard place as many want to switch from their tokens but it will be too hard to diversify the technology.

Breach emphasizes relying on any one system too heavily is poor practice, according to Eve Maler, principal analyst at Forrester Research.

Federal Deposit Insurance Corporation Chairwoman to call on some banks to strengthen their authentication procedures when customers log onto online accounts.

Customers in industries not designated by RSA as being at high risk of attack--and to which RSA may not yet have reached out--will have to pay for the new tokens

Forrester finds many companies looking away from hard token solutions

Poll:Does Lockheed Attack Mean Bankers More or Less Secure?

Sterling Metz — Sun, 05 Jun 2011 19:15:00 +0000

Hopefully everyone saw the recent news about Lockheed Martin has been defending itself against a series of attacks. Back in March it was hypothesized that an "advanced persistent threat (APT)" style attack could only be executed by a foreign power. This attack on Lockheed-Martin seems to defend the claim that the US government and those that support the government are the primary targets. But what about financial targets? Do those in banking security feel confident that the attacks will stop at the government sector or will these attacks will also occur against banking targets.

How does the financial sector feel about this? Let us know in the poll below!

Want to know more about our perspective? Click here to dicuss options beyond tokens!

Secure Identities in 5 Steps

Sterling Metz — Thu, 02 Jun 2011 07:20:00 +0000

STEP 1) Evaluate your activities

Banks interact with their customers in many ways. A structured way to evaluate your activities is to divide all your activities into four boxes along the axis of risk and frequency in order to better see priorities and degrees of focus.

Each set of activities requires a different tool. For example, High Risk / Low Frequency activities need a behavior-driven, two-factor identity approach maybe with a biometric or multi-factor approach with a graduated, account-opening strategy. Aligning tools with processes is essential.

Below is a sample:

STEP 2) Organize your identities

By having a complete view of all applicants and existing accounts, systems get smarter. Strange behavior on an existing account, such as that user’s social security number attempting to open three more accounts, likely is fraud. Only through keying at account opening and managing that key through an account lifecycle can you protect with internal data.

STEP 3) Segment your incoming traffic

Monitor the interactions that customers are having. Did they check their balance, leave, then come back and clear out an account in one payment to a new payee? Has this customer tried to access an account three times today? Fraud and behavior models can help say who is likely to commit fraud to keep the big costs down and the customer experience up.

STEP 4) Put the right tools in the right places

If your strategy is segmenting well, then knowing when to use knowledge-based authentication, when to use two-factor authentication, and when to take no action becomes easy. Model + tool = ROI on business enablement.
Business enablement isn’t very helpful without business continuity. Make sure your High Risk / High Frequency tools come with reliable contingency plans to ensure customers can always access their data. Best of breed two-factor authentication means pass codes through a variety of channels.

STEP 5) Study your results, Adjust, and Optimize

Fraud patterns change regularly. Studying internal results with external analysis can result in tighter and tighter segments, better interaction management, and the safest, most streamlined identity process possible.

To learn more, sign up for our monthly newsletter or follow us on twitter

This post was contributed by:

.

Accelerating Third Party Onboarding And Putting the Aspirin Back in the Medicine Chest

Sterling Metz — Tue, 17 May 2011 08:55:00 +0000

Operations staffing can cause some of the worst headaches. Connecting a network of branch locations and offices securely is tricky without having to deal with third party contractors, temporary workers, and vendors. Managing access as contractors come and go is a cost that seems too high and a process that takes too long. The process is prone to waste: procure a token (shipping, activation), enabling a laptop VPN connector (IT setup, support, debugging) recapture the token on departure (shipping deactivation), and make sure it all works with your identity access management (IAM) system.

Switching to a lightweight tokenless authentication system sounds like a good solution on paper, but what about all those new headaches? Major IAM providers can take 9 months to make such a switch. What about going to a vendor? Some don’t have token and tokenless systems that play well together. Can the vendor scale to my enterprise? Can the vendor integrate to my platform?

Step away from the medicine cabinet. Let’s prioritize.

The first question that has to be answered is how well does this security provider mesh with your multimillion dollar IAM platform. Equifax sees the partnership between IAM system and remote access securer as the biggest asset an authentication partner can have in solving customer business needs. If they mesh with the major vendors, they have a track record of success which means they can scale. Now all you need is a nimble provider who can make it happen. Click here to read how Equifax now integrates with Juniper system and let us know if you want to put the aspirin away.

To learn more, sign up for our monthly newsletter or follow us on twitter

Fight Fraud With More Assets

Sterling Metz — Thu, 28 Apr 2011 00:01:00 +0000

This is the first of a two-part blog looking at the competitive advantage that multiple data set modeling brings to fraud reduction efforts and risk decision making. First, fraud:

Stopping fraud gets exponentially easier with multiple data sets. In fact, they are the single biggest competitive advantage you can erect to fight fraud and identity theft.

Multiple data sets are those data collections which go beyond credit data. They can range from address changes that credit files might not see, to previous addresses, to employment records. By creating the ability to successfully manage multiple data-base results, a company enhances its ability to discover fraud. For example, if an applicant uses an address that your system determines is three years out of date, that would be a red flag. The same applies to employment data that doesn’t match up or if an existing customer trying to extend services has data that doesn’t match.

On the other hand if an applicant can be located consistently on multiple non- credit data sources, more than likely they are who they say they are.
With better fraud flags, you also get to a higher level of real-time “sniff tests.” Your representatives can engage in interactive questions which are set to screen for fraud. For example, in a true name-fraud model, if an applicants can’t remember who provides their natural gas service, they likely aren’t the person they claim to be. In fact, good models will enable you to immediately decline potential fraudsters at minimal cost in either time or dollars.

Using multiple data sets appropriate to your business, you can build a true non-financial risk profile for your products or services. This includes the use of utility and telecom data, bill paying history and multiple address histories. Using these you can quickly identify stolen identities.

It takes a lot of creativity to get to the right attributes out of multiple data sources. Being a data company, Equifax gets to know the customers really well.

To learn more, sign up for our monthly newsletter or follow us on twitter

This post was contributed by:

.

Stolen Data Bring New Risk to Banking Authentication

Sterling Metz — Thu, 21 Apr 2011 00:01:00 +0000

Hackers have recently acquired the keys to a few million castles and it is up to the banks to keep the drawbridges up.

The “keys” in this case are millions of stolen e-mail addresses. Since these now-compromised addresses in authentication, the responsibility to protect against account hijacking has become an even heavier operational issue.

The IDC has posted a great overview of the Epsilon breach and how it impacts retail banks and financial institutions. The e-mail address, a valuable piece of personally identifiable information (PII), is now where real value is for hackers. This is a strong deviation from the actual “account number / PIN Code” theft, which was the sweet spot until recently. This attack is consistent with this year’s Javelin report that highlighted account hijacking as the new golden ticket in fraud.

Phil Hochmuth, Research Manager of Security Products, did a great job of setting the stage for understanding how millions of e-mail addresses have been stolen. Clearly, as Michael Versace, Global Risk Analyst, points out, the biggest danger is that hackers have access to the banking institution and their clients. The combination of which e-mail is associated with what bank gives hackers all they need to launch an aggressive e-mail “spear phishing” and “man-in-the-middle” attacks.

Additionally, e-mail addresses are an authentication factor for many banks. As discussed in our password policies article, passwords are often re-used from site to site. Hackers have the banking institution and the e-mail address, and now just have to snoop around less secure lifestyle sites to get a password. Account hijacking has never been easier!

Versace also discusses how banks are trying to leverage “relationship” based data to authenticate the banking relationship. Some caution is warranted against this strategy. Banks are trying to educate the public about security via “is this your key” and “Confirm this information before logging in” type security. As discussed in our review of EMI vs. Comerica, neither the government nor public opinion is looking to take a caveat emptor approach to banking security. Tech-savvy customers will watch for SSL security and special keys. It can’t be assumed all are that careful. Banks need to leverage device signature or even basic IP address checking to trigger strong two-factor authentication when access attempts are irregular in source. They should check known fraudulent device databases to ensure they aren’t dealing with a known hacker. They need advanced behavioral modeling to ensure that the most obvious account hijacking techniques don’t succeed.

This is the second breach in as many months appear to be more about the value of creating cracks in other infrastructures than the value of breaching the obvious target. The focus for the broader secondary targets is to shore up their front line defenses.

If you want to talk more about improving your fraud prevention, contact a specialist here.

To learn more, sign up for our monthly newsletter or follow us on twitter

This post was contributed by:

.

Locking Out Your Customers Is Ill Advised

Sterling Metz — Thu, 31 Mar 2011 08:37:00 +0000

Online business enablement is one of the chief ways banks are looking to meet the cost saving and customer centricity requirements that will drive growth in the post Dodd-Frank world. Two-factor authentication will play a larger and larger role in making sure there are secure methods for allowing dispersed populations to bank as they need, and do so securely, and at their leisure. Two-factor authentication doesn’t necessarily mean being strapped to a key fob to do your banking. Government guidelines specify “Something,” not what that something has to be. It could be a token or card in your wallet, but wallet and keys get stolen and cards/tokens can be forgotten. Without alternative methods of authentication, customers get locked out. When a customer is on vacation, locked out of their accounts is likely the last place they want to be. Is the ideal customer experience having a customer try to find a friend to go to their home to find and read a six- digit pass code to them while they desperately try to wire money to a troubled relative?

There are software solutions that move beyond tokens. Phone-based systems let customers set up multiple methods of contact, including a cell and land line number. Cell battery dead and away from home? Leverage the “what you are” angle of two-factor authentication and deliver a unique password that has to be spoken as the individual needing to be identified.

Networks, roads, supply chains, and data flows all have systems that sometimes fail. Things break. What separates best in class from run of the mill? Handling the unexpected.

The opportunities for system failures are endless and proper business enablement is rooted in proper business contingency planning. Customer experience shouldn’t be subject to the weakest link in security vendor’s architecture. Leave many roads for customers to get to their money and securely protect them as they move it about.

If you aren’t sure your MFA solution is meeting this need, contact an MFA for Financial Institution specialist here.

To learn more, sign up for our monthly newsletter or follow us on twitter

This post was contributed by:

.

Alternative to tokens for strong authentication

Karthik Mani — Wed, 23 Mar 2011 00:00:00 +0000

The recent news about the breach at a major token based authentication provider has raised concerns with many about the security of their data and how it may have placed their current customer base at risk. Secure two factor authentication has been a standard for strong authentication which creates secure remote access. Tokens have been a big part of that. On the other hand, there are alternatives to token based two-factor authentication that are not affected by this breach. A platform that is configurable at a policy level to turn on and turn off different forms of authentication (tokens, SMS to phones, voice biometrics etc.) provides the ability to weather the breach of one channel.

We acknowledge that this is a very serious situation for all companies providing security solutions, and particularly authentication solutions, to the marketplace. Enterprise businesses have trusted Equifax for over 100 years to protect valuable assets ranging from personally identifiable information (PII) to delivering innovative solutions in fraud detection, decisioning technology, and much more. Equifax has a history of employing best practices for securing data and manage identity, and we continue to review and improve those practices.

Equifax’s Anakam identity and multi-factor authentication solutions were created with federal mandates and NIST guidelines as the guidepost against which we measured our capabilities. We continue to build solutions based on these guidelines and best practices. We provide a single platform that supports multiple authentication capabilities with a configurable framework. This helps business continuity for our customers when faced with these types of situations. We are providing these solutions and professional services to our customers in Government, Healthcare, Financial Services, Telco, Utilities, and other industries. Please contact us here to understand more about the tokenless alternatives for strong authentication.

To learn more, sign up for our monthly newsletter or follow us on twitter

Equifax Provides Secure Multi-factor Authentication

Sterling Metz — Fri, 18 Mar 2011 19:19:00 +0000

The recent news about the breach at a major security company has given many organizations cause for concern about their own data security, particularly when using tokens. Complete data security combines multiple authentication strategies with appropriate enterprise risk security solutions. Security should always include encryption of data, strong authentication combining passwords and second factor authentication, and awareness of policies mitigating possible social engineering. Most importantly, security is based on an awareness of who has access to your data.

Equifax is an innovator in identity proofing – verifying more than 60 million identities annually. Equifax’s Anakam Identity Services offers a comprehensive policy-based platform that establishes trust through variable multi-factor authentication that mitigates risk. This intellectual property is the industry standard and Equifax has led the ongoing development of the authentication of network users, issuance of digital certificates, and Knowledge Based Authentication from a broad set of unique public and proprietary assets.

Equifax’s identity and multi-factor authentication solutions were created with federal mandates and NIST guidelines as the guidepost against which we measured our capabilities. Enterprise businesses have trusted Equifax for over 100 years to protect user data valuable assets ranging from PII and personal user data along with delivering and deliver innovative solutions in fraud detection, decisioning technology, and much more.

If you have questions about the security of your data, please contact us here

To learn more, sign up for our monthly newsletter or follow us on twitter

Future Looking Learnings from EMI vs Comerica

Sterling Metz — Thu, 17 Mar 2011 00:01:00 +0000

The Experi-metal v. Comerica dispute is going to trial and both sides have submitted briefs. The crux of the trial will center on what is “reasonable” security. “Reasonable” is going to require FFIEC-weigh-in. Aviva Litan of Gartner expects that more guidance will come but the ABA’s David Navetta disagreed on the value of what will be offered. Either way the long-term impact could be decided more in the court of public opinion. If small businesses rally together and say that transactional security consulting is something they expect from their bank, financial services will have to keep up. Don’t believe me?

In a survey presented by Doug Dannimiller of Aite, technical integration is a big determinant in which brokerage or banking institution financial analysts and the super-wealthy leverage for their personal finance. Not sure you see the relevance? What is to say that the financial controllers in corporate America have a different set of requirements for who they bank with? Do they use financial systems radically different than financial advisors and super-wealthy investors? Effectively the controller is your chief business user. He will have a voice at the table of what bank is used. Making that role the villain instead of the victim may not be banking’s best move.

The EMI v. Comerica case should give pause on how to approach multi-factor authentication with clients. Comerica let a business client enable unlimited wire transfers based on a single access checkpoint. Should any business be doing 5 or 6 (or 47 in this particular case) wire transfers in a three hour period without reauthorizing. Is that a common business need in any reputable industry?

Behavior-based graduated security is the best practice in all access management. Providing hackers with an opportunity to “phish once and you’re in” creates a problem. Not treating that as a financial institution responsibility creates a bigger problem. The fact this case is going to court indicates a “caveat emptor” approach to e-authentication is not sustainable regardless of what FFEIC says or doesn’t say is “reasonable.”

Equifax recommends enabling multi-factor authentication in online processing of wire transfers from banks to consumers, businesses, or other entities. If you aren’t sure your MFA solution is meeting this need, contact an MFA for Financial Institution specialist here.

To learn more, sign up for our monthly newsletter or follow us on twitter

This post was contributed by:

.

Password Policies and Identity Security

Sterling Metz — Thu, 10 Mar 2011 08:58:00 +0000

In spite of security concerns, many users still choose passwords that are easy to guess, write down their passwords, and share them with others. They also re-use their passwords on multiple sites and use the same passwords on sites that do not house sensitive or personal information, (Social Media, Lifestyle), and ones that do, (Financial Services, E-mails, Retail Shopping) sites. This type of password usage makes the users' identities much more vulnerable to theft and fraud.

There is, of course, another approach to password management that would reduce concerns about guessable or shared passwords. Sites that house or provide access to sensitive personal or financial information could implement two-factor authentication. This would greatly reduce the sites’ vulnerability to password attacks, because without the one-time passcode, hackers could not access the accounts.

The National Institute of Standards and Technology (NIST) recommends different password strengths for Level 2 (user name and password only) authentication, depending on how long a user may keep a password before changing it and how long a user must wait after three failures before trying again. For example, if a web site’s policy is to permit re-try after one minute and to allow users to keep their passwords for five years, a password would need 37 bits of “guessing entropy” to qualify as Level 2. However, if a password must be changed every 90 days and a user must wait a full day before trying to log in after three failures, a 22-bit password would provide Level 2 assurance level.

Since many sites allow e-mail addresses as user names, the only thing between the attacker and account compromise is the password.

Recent studies [link to Microsoft study] on passwords have found that password strength varies for different types of sites. Median password policy strength for dot-com sites is 19.9 bits, 31 bits for banks and other financial institutions, 43.7 for dot-edu sites, and 47.6 for dot-gov sites. Dot-com sites often have less sensitive information to protect and want to keep usability high, so they do not need login complexity. Banks, dot-edu, and dot-gov sites, have much sensitive information to protect and often must meet regulatory requirements for more complex passwords.

Low-hassle out of band two-factor authentication is the best practice for the sensitive information kept inside financial institutions. How are you protecting your customers from themselves?

If you want to learn more about these technology solutions, let us know.

This post was contributed by:

.

Move Past Secrets

Sterling Metz — Thu, 03 Mar 2011 07:52:00 +0000

Five to ten years ago, little known facts about our lives were widely used for authentication before we were permitted access to sensitive information. For example, if someone called a retail bank to check a balance or question a DDA fee, she would provide identifying information for herself and her account, and then would be asked a question for security purposes. Quite often the question was, “What was your mother’s maiden name?” This was a “shared secret” model of authentication. The caller had provided a little-known fact about herself to the credit card company, and the credit card company used that fact to make sure that the caller was who she claimed to be.

Over time, as more organizations use the same facts in various contexts, “shared secrets” become much less secret. As a result, security questions started covering more and different facts. “Who is your favorite historical figure?” “Where was your father born?” “What was the make of your first car?” -- the answers previously provided by the account holder. These types of security questions are still used by many organizations for authentication for sensitive transactions such as password resets on e-mail accounts.

Now, with the proliferation of social networks, genealogy sites, blogs, and other ways for people to disclose more personal information about themselves in more different contexts, there are many fewer secrets than there used to be. Secret questions are no longer good enough as sole means to secure sensitive information. Something more is required. It could be knowledge based authentication (KBA) or a one-time passcode issued to a particular individual and has a limited lifetime.

Such a passcode is randomly or pseudo-randomly generated, so it contains no information about an individual and has no connection to such information. As a result, one-time passcodes can be used for authentication as often as needed without in any way increasing the chances that the next authentication transaction might be weakened because of the data disclosed in an earlier one.
Of course, nothing is infallible, particularly in information security. However, for the foreseeable future, combining dynamic KBA for identity proofing when a credential is issued with a one-time passcode that authenticates the holder of the credential in transactions that follow is a strategy that minimizes the disclosure of “secret” information and increases its usefulness for verifying identity in sensitive environments.

If you want to learn more about these technology solutions, let us know.

This article was contributed by Anna Slomovic. Anna has focused on data protection in industry and government, particularly on protection of personal information. Anna is the Chief Privacy Officer for Equifax.

When Even the CEO of the Largest Social Network Gets Hacked ...

Karthik Mani — Thu, 27 Jan 2011 08:35:00 +0000

We keep this blog focused on sharing best practices and learning in the Equifax focus areas. Periodically there are announcements from Equifax that are relevant to that. Here is an example:

Equifax made significant investments in 2010 in the general area of Identity Management. We realize that assurance of the identity of the person who you are transacting with, when the person is not in front of your employee, is a big priority for Government and Business. Trust is key to move more business processes on-line; trust is also key to reduce the risk in current on-line business processes from ever increasing sophistication of hackers.

Do you want to do business with who the person claims to be?
Do you know that the person is who (s)he is claiming to be?
Can you do this without having to put a physical token on every person?
Can you utilize voice or other biometrics to identify a person without expecting each person to carry around a retina or fingerprint scanner?

Here is the first of those announcements. There are few more exciting announcements in the tap in the next few days. Please check back here or sign up for our monthly newsletter

What are the barriers you see to move more of your business processes on-line?

Fraud Webinar Today!

Sterling Metz — Tue, 30 Nov 2010 00:01:00 +0000

The Equifax Fraud Webinar was today! Thank you all who attended and all who will watch it's replay here!

5 Reasons You Shouldn’t Attend the Equifax Fraud Webinar

Sterling Metz — Tue, 23 Nov 2010 00:01:00 +0000

There is one week left until our fraud webinar! If you’re still having a tough time deciding whether you want to attend, here are 5 reasons you shouldn’t attend:

You aren’t concerned with how current fraud trends are affecting banks – Fraudsters are clever and the fraud economy is bustling. We talked about it in our perfect storm of fraud post, and you can hear more in the webinar.

You aren’t interested in combating identity fraud – Identity fraud has been around as long as ecommerce and the data is getting easier to get a hold of. Has your approach kept pace? We’ll be looking at true name fraud and SSN verification implementations that have helped other companies stem losses.

You want identity fraud to be a cost center on the bottom line – Identity fraud is a necessary evil in a digital age, but your revenue doesn’t have to bear the brunt of the damage. Fraud tools and solutions have evolved so that stopping fraud doesn’t have to stop your business.

You aren’t concerned that your customers’ accounts are being highjacked – What happens when a password is compromised, or when a security question answer is available via facebook? Social media and a fraud economy have made a new approach to customer authentication a necessity for the modern bank.

You like it when customers take their business to your competitors – The modern customer has a short attention span. If your marketing efforts succeed in getting a good customer to apply for a new product but you ask her for five minutes of invasive questioning, she may just walk away.

If you think you SHOULD attend this webinar, click here to register.

Thwarting True Name Fraud Like Never Before

Andrew Skillen — Thu, 18 Nov 2010 00:01:00 +0000

Identity theft impacts over 11 million consumers a year, at a cost to businesses of over $50 billion annually. Many of us are worried about our identities being stolen. Can more be done to protect the consumer? Absolutely.

True-name fraud analytics, packaged together with technology, help provide maximum protection against identity theft. We have identified four data sources that, when combined together, effectively help predict true-name identity fraud:
1) A lenders own data identifying location, time and channel of past fraudulent incidents
2) credit data
3) collections data
4) other internal data (just ask us!)
These data sources can be particularly useful in screening for suspicious activity, and predicting identity fraud. Technology helps streamline the data collection and integration process for feeding this data into the fraud model, and tracking the effectiveness of the analytics. Most importantly, when the model identifies suspicious activity, authentication software deploys out-of-wallet questions against the suspicious inquiries, automating the process of verifying whether a consumer is who they say they are in real-time. Working together, the analytics and technology thwart fraudsters like never before.
A fraud validation will determine the lift you can get from deploying a true-name fraud solution. If you want to learn more about these predictive data sources, analytics, and technology solutions, let us know.

Are there other assets you use in fraud modeling? Do you have a different approach to true name fraud? Leave us a comment!

Know, Catch, Keep Presents an Identity Fraud Management Webinar

Sterling Metz — Tue, 16 Nov 2010 07:06:00 +0000

Identity Fraud trends point to higher impact on the bottom line for banks. Intrusive efforts to prevent identity fraud are a serious problem for customer centric organizations as they are time consuming, frustrating, and ultimately off-putting for customers.

Banks are focused on customer centricity while minimizing write-offs. Identity fraud is the area that drives these business objectives in opposite directions. While balancing these two business goals, what if you could manage 70% of your identity fraud while only reviewing only 10% of your traffic?

Identity Fraud Management that combines the best of fraud analytics, authentication technology, real time business rules, and back-end business intelligence can help banks stay in step with federal regulations, lessen the impact of fraud losses on revenue goals, and allow more good customers to flow through in less time with less interruption. The right combination of acquisition segmentation models, social security number verification, and graduated business rule enabled tools can mitigate the cost of preventing fraud and create the optimum point for meeting customer service and revenue business goals.

Click here to register for a one hour webinar that will review the current trends in the fraud space, discuss the best practices to best combat the growing ID fraud trends, and review results from successful organizations.

Perfect Storm of Fraud

Rich Huffman — Thu, 11 Nov 2010 00:01:00 +0000

When going through recent financial news, I became overwhelmed at the fraud issues I saw. I came across this article on the cost of stolen cards to fraudsters. When combined with this study on fraud losses, and the 2010 Javelin report on individuals foregoing identity fraud prevention, we could be witnessing the golden age of the fraud economy. Good attacks, cheap identities, low public interest? No wonder banks pay three dollars for every dollar stolen!

Combating this storm is going to fall on the bankers. It’s time to look at fraud solutions. The first concern is that with the kind of data available in Kreb’s article (Date of Birth, Maiden’s Mother Name, etc). ), how how is your account management process insulated? Nothing compromises the industry standard challenge questions like alternative data sets. If you aren’t using a multi-factor authentication tool, then you are relying on the same questions that www.rock3d.cc is answering for a mere $2.10! Multi-factor can take the routine data gathered to provision an account and get back to the real account holder. Wiring an entire balance out of an account? Opening a second and third credit line? Changing the challenges? These aren’t common activities and customers will be reassured by the added security.

The second concern is that the information stolen will be used not just for retail purchases. If a stolen ID has enough credit to establish the card that was just purchased, why not open a few more revolving tradelines? This is where a graduated fraud policy protects from fraud loss. Protect against a valid SSN with a fraudster’s name and address with a synthetic fraud solution. If the fraudster tries using the information to commit true name fraud, challenge not with compromised “In wallet” questions that can be answered via Trojan data. Get questions based on the rarely thought about “out of wallet” data. The process is more invasive but if you have optimized your approach, you will get the maximum bang-for-your-buck while minimizing customer impact.

Individuals are pinching purses on fraud prevention while the Zeus attacks are feeding the global ID theft economy. Knowledge Based and Multi-Factor Authentication solutions inside of a comprehensive, progressive fraud policy keep fraud costs down and keep customers trusting in the largest growth channel of banking.

KBA and MFA – The Alphabet Soup of Authentication

Rich Huffman — Thu, 04 Nov 2010 00:01:00 +0000

We talked about the benefits of a fraud model and the bad practices that don’t include them. All of the approaches discussed ended in an authentication solution. Choosing the right authentication solution depends on where in the customer lifecycle authentication is needed. For loan origination fraud solutions, Knowledge Based Authentication (KBA) is king. KBA is a combination of real and synthetic questions designed to challenge fraudsters. Many existing KBA solutions ask questions found on driver’s licenses and other “in wallet” data sources. Javelin Strategy and Research has reported that over 79% of ID fraud is based on a stolen wallet, checkbook, or over the shoulder password theft. If the fraudster has the wallet, the questions need to come from somewhere else. The solution is to find information not stored in public record. Developing a question set that is both predictive and not overly invasive from that data creates the best KBA solution.

KBA is useful at point-of-sale, but what about your existing customers? Imagine they just want to log on to their account and can’t remember their challenge question, or maybe they are trying to make some changes to their account information. Does 5 minutes of questions really seem like a customer-focused approach when they are just trying to get their account back? Multi-factor authentication (MFA) fraud solutions can help. MFA refers to using at least 2 different methods of identifying someone. The recognized practice is to choose at least two of the following three:

• Something you know (a password)
• Something you are (a thumbprint)
• Something you have (a token or cell phone)

Asking someone for two different passwords doesn’t count; that’s just “something you know” asked twice. You can’t really substitute a biometric test because few people can provide that information from their workstation. The traditional multi-factor approach has been “something you have.” Tokens are easy to use but aren’t cost-effective in the mass market. An emerging best practice is a light-weight SMS text. Banks can leverage existing customer data (phone numbers) to allow clients to engage in online activities while still being confident in their client’s identity.

Whichever business problem challenges your fraud solution, make sure you have the right tools to meet the need head on.

If you want to talk to an Equifax specialist about your authentication solution you can e-mail us here

If you are interested in a monthly e-mails about blog updates, sign up for our newsletter

Rich Huffman is VP of Product Management for Equifax Technology and Analytical Services. Rich is responsible for driving the market strategy for Equifax's portfolio of ID Verification and Authentication products.

Rich brings over 17 years of experience in product management to Equifax. Prior to joining Equifax, Rich created and managed products for the financial services industry at Harbinger, S1, and ADP. Rich is an expert in utilizing ID verification and authentication technologies in addressing Red Flag and other compliance concerns for online banking and online payment related activities.

Rich graduated from Clemson University with a B.S. in business with a concentration in economics.

Do You Need A Fraud Model?

Rich Huffman — Thu, 28 Oct 2010 00:01:00 +0000

We reviewed the nuts and bolts of how fraud models work and how they can provide your solution with lots of information. As we discussed in our identity fraud solution cost benefit analysis, many risk analysts still see fraud models as a waste. Let’s discuss some different approaches commonly used in place of a fraud model. The two most common approaches we see are:

· Running an authentication-only fraud solution after pulling credit data

· Triggering an authentication fraud solution based only on internal data

Approach I – Running an authentication-only fraud solution after pulling credit data

Advantages – The credit report is a required element of any credit request, and credit bureaus can append useful elements to help with the fraud determination.

Disadvantages – The primary disadvantage is cost. Credit reports are the most expensive part of a credit request. Pulling the credit report only to discard it is wasteful spending. A fraud model can give all of the same output that commonly gets appended to a credit request without the costly credit data. Another disadvantage is the effect on the customer experience. Relying on your credit pull as the sole fraud indicator will create many false positives that will require identity authentication. The invasive nature of this process will push away profitable customers unwilling to deal with an additional 15-30 minute headache.

Approach II – Triggering an authentication fraud solution based only on internal data

Advantages – Internal flags like channel and a possible promotion code are great indicators of potential fraud. Our analytics team has reported impressive gains when this data is included in modeling efforts. Internal data also serves as the cheapest form of segmentation, as there is neither a model cost nor a credit pull.

Disadvantages – Internal data doesn’t present a complete picture. It doesn’t benefit from fraud “hot lists” or other information managed by third party providers. It also doesn’t benefit from velocity monitors that catch cross-channel fraud. It lacks the predictiveness of a fraud model. A fraud model has all of these characteristics and serves as a better segmentation tool than internal data alone.

Best practice fraud solutions need a fraud model. What is your approach, and how does it compare to our examples?

If you want to talk to an Equifax specialist about your fraud solution you can e-mail us here

If you are interested in a monthly e-mails about blog updates, sign up for our newsletter

Rich Huffman is VP of Product Management for Equifax Technology and Analytical Services. Rich is responsible for driving the market strategy for Equifax's portfolio of ID Verification and Authentication products.

Rich brings over 17 years of experience in product management to Equifax. Prior to joining Equifax, Rich created and managed products for the financial services industry at Harbinger, S1, and ADP. Rich is an expert in utilizing ID verification and authentication technologies in addressing Red Flag and other compliance concerns for online banking and online payment related activities.

Rich graduated from Clemson University with a B.S. in business with a concentration in economics.

What Have You Done For Me Lately?

Sterling Metz — Wed, 20 Oct 2010 10:47:00 +0000

As budget season begins to set in, banks across the planet begin quantifying "what has my vendor done for me lately?" With the BAI conference, the key bank officers are going to see hundreds of vendors positioning themselves for attention. We all know we have to drive value year after year to keep risk, marketing, and technology officers at the top of their industries. We are monitoring the vision of our industry. Just the other week one of the world’s largest financial institutions publicly discussed the need for dynamic partners with a vision for organic growth. Financial technology companies will be challenged to innovate for the sector. Which vendors can banks rely on to improve wallet share with their customers? Will the vendors provide the identity solutions that help empower web strategies? The financial sector needs customer focused technology from trusted partners.

Equifax has taken the downturn to evaluate our approach as well. We heard that understanding “ability to pay” was a requirement of the new normal, so we focused our analytics to help better understand actual income. We understood that wallet share was a growth initiative, so we advanced our master data management to establish complete understanding of the customer. We saw growth in the desire of banks to empower their customer to manage their instruments through the online channel, and we acquired Anakam to position ourselves as a company focused on the end to end security required for you to execute those goals. We have a roadmap around customer lifetime value and identity fraud management to keep us delivering the goals that support your growth initiatives going forward.

Bank Technology News and Financial Insights think we are on the right track. For the fourth straight year we have ranked in the top 25 on the FinTech 100, something no other information solutions company has achieved.

InformationWeek recently put Equifax at 29 on its top 500 list. Looking deeper, we ranked as the best vendor for technology innovation.

And while all of this momentum has been great, it is your business that we are most focused on. To sustain this momentum, we need your feedback. Has Equifax innovation been aligned with your business needs? Let us know.

Leveraging a Fraud Model

Rich Huffman — Thu, 14 Oct 2010 08:42:00 +0000

What is the makeup of a fraud solution? We’ve reviewed why you need end-to-end fraud management, but let’s make sure we explain how all the pieces work together.

Fraud Model – A tool designed to help segment your population and determine the likelihood of fraud. The customer logs onto www.yourhomepage.com and says she is looking to open an account. By asking her for basic personally identifiable details, a fraud solution can ask a lot of questions. Am I getting an SSN? Is this a real SSN? Can I link this SSN, name, and address to some externally verified data source? Is there anything suspicious about this transaction? How many times have I seen this SSN recently? Is this the third time in a week I have seen it through online channels? Has this SSN recently been reported as a stolen identity? The earlier you introduce a fraud model into your loan origination process, the quicker these red flags can be set off.

Half of those questions indicate fraud that no bank should entertain. If too many of the questions have the wrong answer, stop processing! Don’t try to extend an offer of credit, don’t consider cross-sell, and certainly don’t waste time and money asking her an invasive set of questions she won’t be able to answer. If a potential client customer can’t give you a name and social security match, is she going to know who provides her car insurance?

The data produced by a risk model can speak volumes to a risk analyst. Ultimately your risk analysts are paid too much to handle these credit requests individually and should only be managing process. Rule flows should determine how every question answered by the model impacts the credit request. Is it a real SSN with no history of abuse but the address is a prison? Perhaps this request needs to flow to an authentication solution. Is this a valid identity but the SSN has been used four times through your online channels today? Maybe this should flow to a call center representative. A good fraud score and reason code driven process creates the most information early on in a request’s lifecycle, allowing for the most tangible benefit later.

Stay tuned later this week for more details about how fraud model outputs can be used downstream.

If you want to talk to an Equifax specialist about adding a fraud model to your solution you can e-mail us here

If you are interested in a monthly e-mails about blog updates, sign up for our newsletter

Rich graduated from Clemson University with a B.S. in business with a concentration in economics.

"Next Best Action” vs “Next Best Offer"

Chandresh Modi — Thu, 30 Jun 2011 00:01:00 +0000

Winning the attention of customers that are over-marketed and suffering from “ad-fatigue”.

We live a world where companies are attempting to deepen relationships with existing customers in order to sell products that are most likely to be consumed. Terms like “cross-sell”, “increased share-of-wallet” and “right offer, right customer, right time” are the popular themes in a world where retail banks strive for new revenue growth. The aforementioned terms all scream “next-best-offer,” however, often all that’s achieved is the refinement of a sales pitch, not ad-fatigue, so you need to be careful when targeting customers.

Ad-fatigue occurs when customers are inundated with irrelevant marketing and/or sales pitch at each inbound/outbound interaction point. When was the last time you received mail campaigns in the post and actually read them? If the answer is never, then you have been induced into the comatose state of “ad-fatigue” through over-marketing. This makes it increasingly difficult for a retail bank to get their message through to a customer. To solve this we need to get smarter.

“Next best action” succeeds where “next best offer” fails by first ascertaining what type of action is best at a specific point. While it is important to ensure that an offered product is relevant to the customer’s needs, it is equally important to know when not to offer a product, rather do something from a service standpoint to increase brand loyalty.

Examples of actions that can be taken;

It is important to note that customers do not react the same way when targeted. One set of customers will accept your offer while others will completely ignore it or worse, react negatively possibly causing attrition. “Next best action,” done right, creates brand and customer loyalty while minimizing customer attrition.

To learn more, sign up for our monthly newsletter or follow us on twitter

This post was contributed by:

.

Who Let The ("Sleeping") Dogs Out?

Chandresh Modi — Thu, 23 Jun 2011 00:01:00 +0000

What are ‘sleeping dogs’ and why you shouldn’t market them

Today’s cross-sell efforts are focused on deepening customer relationships. Most companies target existing customers using traditional response modeling to determine responders vs. non-responders to find which customers to treat. However, this somewhat antiquated method of targeting existing customers can come with some serious repercussions.

Uplift, or incremental response modeling, is an analytical technique that helps identify what ‘purchasing’ state the customer is in. This technique is based on the fundamental premise that each customer, whether treated or not, will behave differently. Some customers actually will buy what you’re offering whether you treat them or not. These are your “Sure Things”. Customers who will only buy if treated are “persuadable” and this is where your resources and marketing dollars must be focused. “Lost causes” will not buy your products, and “sleeping dogs” often will defect if treated. To visualize, your customer falls in one of four boxes for every product at any moment in time. It is your job to know where they fall to know if you should treat them or not.

Treating someone considered a sleeping dog not only can discourage the customer from buying, but worse, bring focus to the entire banking relationship and cause the customer to churn. The very action designed to deepen the customer relationship can create an adverse reaction. Don’t worry, an individual identified as a ‘sleeping dog’ doesn’t always remain in this state. Factors that drive “right customer, right offer, right time” determine whether a customer is considered a sleeping dog or not and this state can change over time.

Traditional marketing and offer management strategies determine success simply by measuring offer response rates and set the bar so low that a measly three percent response rate is deemed successful. The effectiveness of marketing and offer management strategies should be designed and measured based on the positive or negative impact treating a customer can bring. Only then do you see the true benefit of these strategies and uplift modeling most effectively identifies targets and determines success.

To learn more, sign up for our monthly newsletter or follow us on twitter

This post was contributed by:

.

The Offer Management Spin on Fairbank's Earnings Call

Sterling Metz — Fri, 10 Jun 2011 06:54:00 +0000

American Banker reported some remarks from Capital One CEO Richard Fairbank we find relevant to risk and marketing officers and worthy of comment.

“Fairbank said issuers have been richening rewards inducements and that account origination costs had become high. He predicted that cardholder attrition would rebound after a period during the recession when consumers "wanted to hang on to what they have."

In addition to the obstacles posed by heated competition, Fairbank was cautious about the overall environment for loan growth.

He said that while household deleveraging is stabilizing, "I don't think the consumers [have] fully demonstrated how much collectively they are going to step up and be the same consumers they were before with respect to cards."

Regarding the high cost of account origination, Fairbank cites rewards as the driver. Rewards may be creating a new problem, but we suspect marketing is still the bulk of that cost. Marketing costs get divided over each new origination. Driving better offers through in-portfolio cross-selling can reduce marketing costs by a factor of five. Outbound marketing campaigns targeted to potential customers with ability and propensity to accept would help as well. Equifax UK has found that over 30% of marketing offers are declined after interest. This means wasted marketing, risk, and operations dollars.

Regarding increased cardholder attrition, we feel more options will mean more attrition as credit losses continue to fall and risk policies allow for better offers. Stickiness will be key to keeping your balances high. Increasing share of wallet through in-portfolio cross-selling will ensure it is your cards that stay in your customer’s wallet. Reducing APR or DDA fees because of multi-product relationships will reduce churn and prevent defection.

With regard to increased credit spend, we can only say, “My, what short memories American consumers have.” Under Fairbank’s rosy outlook, increases of a few basis points here and there on the expense side, or a few more in the loss column, will be hidden in the bottom line. We would recommend maximizing spend with offers that increase lifetime value across all lines of credit in your institution. Lower-cost acquisition and better retention will create the greatest benefit from greater customer spend.

The goals of Equifax’s strategies are protecting your customer acquisition strategy and maximizing customer value over the long haul. Click here to read more.

To learn more, sign up for our monthly newsletter or follow us on twitter

This post was contributed by:

.

Durbin, Dodd, and Frank; Why Senators Won’t Be Vacationing In New York, Charlotte, or San Francisco

Sterling Metz — Thu, 14 Apr 2011 00:01:00 +0000

As a result of Congress’s financial reform efforts, two storm clouds of increasing costs are looming on the retail banking horizon. First is the cost of cleaning up “too big to fail.” Specifically, the FDIC’s Sheila Blair says new “living will” programs required by the Dodd-Frank bill may require “significant and potentially costly restructuring.”

The other is increased costs from churn that likely will result from the loss of the consumer’s favorite service – free checking.

Free checking, the here-to-fore best marketing tool for customer acquisition looks to be doomed by the proposal to cap debit card fees. The ABA is reporting that a survey of bankers indicates two of three banks will remove free checking. The LA Times reported the beginning of the end in early of February. A couple of banks have been denying they will remove their popular entry-level product, but the ABA didn’t say elimination was the only option. Some 81% will increase current fees and / or remove their debit card reward program. Almost half said they were considering debit card transaction fees. Impacting demand deposit accounts and free checking services will create even more churn, the question will be how much and who will be affected.

In addition to the inherent costs in replacing customers, churn also creates a serious problem for cross-selling. Since it’s easier and cheaper to market to existing customers , it’s vital to limiting churn as much as possible. Key to this is knowing who you hold on to and who you let walk. To make this decision, a bank needs to see the customer from more than one angle.

A single view of customer will make sure you see the auto loan and the two mortgages that the other sides of the house want to protect. Adding wealth data and advanced behavioral analytics will make sure you are taking the actions that lead the customer to consolidate their assets at your institution.

Retail banks are going to be challenged to take the next best action for a the portion of their customer base that is content in keeping their money spread across many institutions. Make sure your next step is your best step.
The goals of Equifax’s churn prevention strategies are protecting your customer acquisition strategy and maximizing customer value over the long haul. Click here to read more.

To learn more, sign up for our monthly newsletter or follow us on twitter

This post was contributed by:

.

"Customer Centric?"

Chandresh Modi — Thu, 07 Apr 2011 00:01:00 +0000

Highlights from Best Practices in RETAIL FINANCIAL SERVICES Symposium

Financial institutions are at a cross road where cutting cost is no longer an option. In light of that reality, the themes that resonated most at the 16^th annual conference centered on customer centricity, customer loyalty and the phenomenon of social media.

Is customer centricity a cliché?

“Customer Centricity” is a phrase that is slowly becoming a cliché since there is much more talk about it than action. The concept is not a new one and has been around since the decade-long attempts at implementing CRM systems. So what has changed? Why is it different this time? And why must it be successful? The fundamental difference is the regulatory environment post the collapse of Lehman Brothers in 2008 and the fall-out from bank closures. Terms like “Dodd – Frank” and “Durbin” make all bankers cringe. The impact of such legislation could cost the industry upward of $37 billion in lost revenue according to one presenter. One bank highlighted that the ‘Durbin’ legislation would render 80% of its deposit customers unprofitable.

Filling the large revenue gap is critical and can only be achieved by focusing on existing customers and deepening penetration through a level intimacy not yet exhibited. Technology and Analytics play a big part in achieving customer centricity in the following ways:

-Implement a 360 degree view of your customer
-Visualize your portfolio and key performance measures
-Operationalize your “next-best-action” and “segmentation” strategies
-Deploy a multi-channel and multi-product real-time decision support solution
-Deploy real-time decision support solutions that integrate with your existing enterprise systems

Legislation will make banks think about customers with which they want to deepen relationships (many banks at the conference cited ‘mass affluent’ as a key target segment) and those that they are happy to let attrite.

Brand is critical for customers when choosing to buy, stay or leave

Customer loyalty, critical to business strategy, is largely being driven by changes in the regulatory and economic climates and is being empowered by the advent of social media. The motivation for this shift is the reality that deeper share of wallet can only be achieved through customer loyalty.

The challenge for banks is moving their respective sales organization, which has historically focused on being the best sales organization, to one that is the best customer-centric organization. This is not to say that you should ignore sales altogether, but rather to have a balanced approach that focuses on the customer’s needs.

Think of each customer as a brand ambassador. Positive brand recognition creates brand affinity which in turn generates customer referrals. One bank cited that 30% of new accounts are through referrals. Conversely, each lost customer is an ambassador of bad news that can negatively impact your brand. Research has found that an upset customer tells on average eleven people about an unhappy experience and those eleven people tell five others(source: Customer Win-back, Griffin-Lowenstein).

Technology plays an interesting role in customer loyalty. Customers are intrigued by technology given the explosion of broadband and mobile technologies. The more a bank can enable their services through technology, the greater pull they are going to have in driving customer loyalty.

The innovation engine that is social media

Social media keeps getting more and more attention, but the challenge is “should you decide to monetize it?” and, if so, “how?” Many banks are leveraging social media to stay engaged with their customers, address complaints and enhance their brand. On the other hand, introducing a product or a sales approach through social media could have an adverse effect and the active feedback gained helping customer loyalty and product design might start to disappear. The benefits of a vehicle that can serve as an innovation engine driving differentiated products start to get sparse.

Statistics below highlight the engagement opportunity of social media for financial institutions

61% of online adults participate in social media¹
23% of adult online time is now spent on social networks²
36% of online adults with $100k+ annual income read or subscribe to financial blogs³
19% of online adults have become a fan, friend or follower of a financial institution³
30% of small business decision makers have used social media to research banking providers in the past year⁴

Retail banks are at an inflection point. Headwinds caused by the regulatory climate are causing banks to rethink the way they make money and opportunities lie in creating deeper more profitable customer relationships.

¹ Emarketer, 2010;
² Neilsen NetView June 2010;
³Synergistic Research, American Banker;
⁴EMI Small Business Survey, November 2010

To learn more, sign up for our monthly newsletter or follow us on twitter

This post was contributed by:

.

Modeling for Change – Multiple Model Methodology

Sterling Metz — Thu, 24 Feb 2011 00:01:00 +0000

Besides vintage analysis , Our previous article mentioned that other scores can be useful in managing credit in uncertain economic times. The questions a retail bank risk analyst has to ask is, “What are the combinations that make sense? How do I make the risk score smart again? How am I going to actually make it happen?”

Bankruptcy scores have helped major banks understand their prescreen solution. By combining real income data with demographics and forecasting, Equifax has put our advanced analytical ability into one of the better income models out there. Ability to pay is being seen as a good combination with risk, and blending the scores effectively is essential to determine the value. A study in support of our Premium CPR data product indicated that having a job was a strong indicator of ability to pay. For determing the risk of offerings where the ability to pay isn’t the issue, deep dives into customer payment history can yield results in determing propensity to pay. Do customers in segment X pay mortgage or credit card first? What about phone? Dedicated analysts raise the level of the risk debate when they think beyond just the credit score, especially given it’s questionable use as a sole predictor in the current economy.

If analysts empower analytic professionals to leverage a decisioning platform capable of orchestrating multiple data products, credit risk can be managed much more efficiently. The right tools make the analytics talk by pulling multiple scores and indicators, comparing them to an adjustable scoring matrix, and delivering predictions to reality.

The resounding question from all of this is “I only have X analysts and they are focused on driving my EXISTING performance.” Turbocharge your team, then. An external group of analysts who focus on these products augments your results. They have experience with the data and with the calculations because, if you have the right team, they built the products that are now being blended. Custom decision matricies have a way of turning out better when you let your chefs shop for the grocieries. Want to talk to an Equifax Analyst about better refining your scoring in the new normal? Send us an e-mail..

If you are interested in a monthly e-mails about blog updates, sign up for our newsletter

Click here to find out more about Equifax's advanced analytic insights.

This post was contributed by:

.

Vintages Not Known by a Sommelier

Sterling Metz — Thu, 17 Feb 2011 00:01:00 +0000

Altering credit strategies for maximum accuracy has to be the task of credit managers in the current environment. In our analytics leader's final post, he talked about what are some steps that can be taken. His best point was “Am I doing all that I need to be doing to accurately determine risk management policy, considering the reality of economic changes?” He mentioned doing vintage analysis on younger accounts as a good way to track bad rates. The news has blessed us with a perfect example of when this method might help drive policy change.

Right now, no one has missed the news about the iPhone and it’s big relaunch party in February. Expectations of migrations ranged from 11 to 13 million iPhone users in the first year. Obviously this number was to be heavily front-loaded towards the launch. How many conversations do you think Marketing and Risk had about the acceptable risk of customers given the size of the launch? What about credit operations? What was the approach? Toy with the score bands? Loosen the criteria? Prepare a resume and a letter of resignation? Credit score predictiveness is determined after 24 months of performance. 24 months ago people were just figuring out what “apps” were. Timothy Geithner had just announced the US government was in the business of purchasing toxic assets. Do you think starting from that period in history accurately forecasts the bad rates on 11 to 13 million new phone activations?

The case in point focuses on why there is a need to track and compare delinquency levels month by month. The question though, is how. How does a retail banking risk analyst team understand this? It starts with data visibility. Understand that vintage analysis requires data to be aggregated. Whether directly dumped to a group of analytical experts or fed into a data warehouse capable of driving actionable vintage KPI’s, actionable advanced analytic insights start with centrailized viewing of the data. A good team or tool can track accounts as they activate and provide relevant one, three, and six month performance stats that can indicate if policies are accurate. When combined with adequate decisioning frameworks, analyst predictions can drive policy change.

To get back to our example in question, by my count there are 19 different mobile phones that could serve as voice and data predictors from 2011. Do you think any of those devices map better to future trends than bad rates that are being reported now? Delivering on these results requires a trusted squad of internal analysts with a system capable of vintage reporting and segmentation. If you want to talk to our team of trusted squad of analysts with their systematic vintage reporting and segmentation system, let us know.

If you are interested in a monthly e-mails about blog updates, sign up for our newsletter

Click here to find out more about Equifax's advanced analytic insights.

This post was contributed by:

.

Managing Credit Portfolios Part 3

Sterling Metz — Thu, 10 Feb 2011 00:01:00 +0000

Using the full range of tools

Ignoring macroeconomic changes will result in greater inaccuracy of bad rate projections over time, and therefore unexpected consequences for portfolio delinquency and sub-optimal pricing strategies. While there is clear value in macroeconomic data for predicting portfolio performance, each organization needs to implement the approach that best meets their needs.

Other approaches also can be effective at improving accuracy as well. Thus, each of us needs to compare what we are doing today to the full range of advanced analytical approaches and tools available to determine the best way forward.

The first step is to evaluate whether we are doing everything we can to maximize the accuracy of traditionally generated forecasts. Some examples to consider include:
1. Frequency of Forecast Update. Most people recommend annual portfolio review and odds chart update of a scorecard. However, what would the benefit be of updating more frequently? Using a percentage error based calculation (WMAPE: weighted mean absolute percentage error), we see a 16% reduction in error when moving from an annual to a quarterly update. Conversely, if you don’t even do annual updates, the consequences can be dire. For example, when using a four year old forecast, the error increased by 61% over the annual error rate.

2. Using Other Scores. If we are trying to create the most accurate assessment of risk associated with an applicant or account, are we looking at all the information available to us? A risk score alone does not tell the whole story. Other possibilities include affordability scores, bankruptcy scores and profit measures. Just as a balanced investment portfolio mitigates risk associated with each investment, balanced decision criteria can mitigate risk associated with each behavior forecast.

3. Vintage analysis on younger accounts. One of the problems with a risk score is the time horizon used to forecast behavior. If a score forecasts bad rate after 24 months, then you have to wait 24 months to see how accurate the forecast was. Meanwhile, decisions are being made without any understanding of how the bad rate may be shifting. But there are other options. By tracking and comparing delinquency levels month-by-month for each month’s (or quarter’s) new accounts, an early read on delinquency changes can be seen. This vintage analysis will allow for a cut-off strategy change in the right direction, but the size of the adjustment will not be precise.

We saw a 16% reduction in WMAPE by moving from annual to quarterly traditional updates. By incorporating macroeconomic data, as discussed in Part 2 of this series, Equifax achieved a further 27% reduction in WMAPE. This indicates opportunity for significant error reduction with the addition of macroeconomic data.

Conclusions

Major errors in forecasted portfolio bad rates can be avoided. The question every risk manager needs to ask is, “Am I doing all that I need to be doing to accurately determine risk management policy, considering the reality of economic changes?” If you are expecting a changing lending environment, whether it be associated with changing consumer behavior, changing lending practices, or macroeconomic changes, then are you using all the appropriate data and tools available to manage the changing environment effectively? The tools are there. Embrace them.

If you are interested in a monthly e-mails about blog updates, sign up for our newsletter

Click here to find out more about Equifax's advanced analytic insights.

This post was contributed by:

.
Vice President Modeling and Analysis

Real World Adventures in Offer Management

Sterling Metz — Tue, 08 Feb 2011 00:01:00 +0000

“Maximize the long term profit potential," "optimize your customer lifetime value," "deliver profit through deep customer relationships." These are some of the new wave of business mantras being adopted by players in all industries. A top 50 company needed to rethink its strategy coming out of the wild west of its industry's massive restructuring. A goal was developed to build common cross channel strategy. After consultation with Equifax, Marketing felt they needed the ability to support the selling of the right product to new and existing customers across all inbound and outbound interaction channels. Risk needed to reduce write-offs across their portfolio. Both needed cross channel decision support delivery of offer management to their front end systems. Equifax delivered a framework leveraging our "capture, predict, optimize, and learn" philosophy.

Step one: Capture - Getting their internal IT to bring together over 30 internal data sets wasn't going to happen quickly, so Equifax created and managed a data warehouse providing a repository that enabled a 360-degree view of the customer. This lets us deliver the key insights necessary to build behavior trends. For example:

Jon Smith has products across three lines of business but doesn’t cover his fees on one of them. Jane Doe has had good payment history and credit data to indicate she has room for more but how much more?

By managing enterprise-wide portfolio and behavioral data and augmenting it with external descriptive data, the customer was able to generate a complete, single customer view.

Step 2: Predict - The solution determined the characteristics indicative of one's ability to pay, willingness to pay, and desire for more service. These characteristics were combined into custom decisioning models capable of predicting the buying and risk potential of the consumer. These factors combined to create insights into "what will happen next."

Step 3: Optimize - Through a flexible decisioning framework, model output analysis could change and drive action depending on the needs of the business. Propensity to buy and expected benefit to the customer drove the ongoing optimization effort. Event and time driven triggers ensured account level analysis was happening as necessary to distinguish "what's the best that can happen" and drive intelligence into action.

Step 4: Learn - Learning is a critical aspect of the solution. This feedback loop enabled the customer to adapt to changes in market conditions. Providing an actionable intelligence solution was critical. The solution's ability to learn was composed of two elements:

Continual data refreshes let Marketing and Risk Management see the effect of their banding decisions on their portfolio. Were product counts per consumer increasing? Were write-offs coming down in targeted populations? Only with an ongoing data loop could these insights drive adjustment.

Visualization capabilities provide insights that drive intelligence into action. Leveraging dashboards, ad-hoc reporting, and cube analysis, empowered risk and marketing analysts to accurately identify root cause and develop strategies to positively impact their portfolio.

The Equifax solution provided a central view of customer, a flexible, analytically-driven decisioning service, and an actionable intelligence environment. In small test call centers, offer acceptance doubled as existing consumers could be adequately sold to. Marketing, prescreen, and account review regulations were managed with little to no thought from our customer, which was able to meet its offer acceptance and write-off goals from a single technology and analytics partner.

If you are interested in a monthly e-mails about blog updates, sign up for our newsletter

This post was contributed by:

and .

Managing Credit Portfolios Part 2

Sterling Metz — Thu, 03 Feb 2011 00:01:00 +0000

Introducing Macroeconomic Data

Lenders have changed their lending criteria, regulators have been encouraging portfolio shifts away from sub-prime, and, it appears consumers have shifted their behavior. All will be related to economic changes in some way. It makes sense to see how much of the changing bad rates can be explained by macroeconomic factors, and how much is left unexplained that must result from other factors.

One of the problems with traditional bad rate projections, is that they are based totally on historical information. With macroeconomic data – for example, personal expenditures, residential housing rates and savings rates -- we have past data as actual fact, but we also have future projections that can be incorporated in our bad rate projection analysis by leveraging some advanced analytics. This “forward looking” view has the potential to reap great benefit.

To assess this, Equifax looked at historic macroeconomic data and projections, using relevant variables. The models predict the bad rate on auto loans within 18 months of the application date. Figure 3 shows an example of a model built in September 2009, predicting performance over the following 18 months. The model predicts bad rate with good accuracy, even predicting turning points in bad rate (when the trend changes direction)mid-way through the 18 month period. The blue line shows the predicted bad rate from models using macroeconomic data at the application point in time. The white circles show the actual bad rate.

Figure 3: Bad rate projections based on macroeconomic data

Applying this approach to an 11 year period back to 1998, we observe a predicted bad rate that generally follows the actual bad rate. We even see turning points accommodated well. However, while many of the time points have a remarkably good match of actual bad rate to predicted bad rate, there are several points in time where the fit is not so good. Examples are early 2002 and late 2003. These were times when a changing bad rate was accurately forecasted, but the timing was off by six months, indicating either a delayed or a faster response by consumers to changing economic circumstances. This is consistent with consumers sometimes taking awhile to respond to actual economic changes, while at other times their changing behavior contributes to macroeconomic changes.

While macroeconomic variables can have a very positive effect on predicting delinquency, there are a number of more simple predictive techniques that can be applied in the short term to enhance accuracy. We will discuss some of these in part three of our series, and see how they compare to the improved accuracy that results from incorporating macroeconomic data in the delinquency projections.

If you are interested in a monthly e-mails about blog updates, sign up for our newsletter

Click here to find out more about Equifax's advanced analytic insights.

This post was contributed by:

.
Vice President Modeling and Analysis

Managing Credit Portfolios in Changing Economic Times Part 1

Sterling Metz — Thu, 27 Jan 2011 00:01:00 +0000

Better Delinquency Forecasting Stems Losses
The Business Issue

Due to the radical shift in the economy, financial institutions have suffered significant portfolio losses in recent years, primarily due to an inability to accurately forecast bad payment rates. Specifically, many lack advanced analytic forecasting tools that successfully associate macroeconomic factors with delinquency rates.

For example, let’s suppose it is June 2008. An auto lender accepts loans with a score of 695 or higher. The most recent bad rate projections show a 0.51% bad rate for the accepted population which would equate to a projected loss of $36MM. The actual bad rate ends up being 0.98% for these 2008 loans. For 1 million loans, with an average write off balance of $7,000, and all the bad loans go to write-off then the actual write-off balance would be $68MM. And, if the lender set risk policy based on an older bad rate projection, the error would be even larger.

This clearly illustrates that economic changes and associated changes in consumer behavior have not been accommodated and managed accurately; otherwise these unexpected – and devastating -- losses could have been avoided.

Did the problems result from poor loss-forecasting techniques, or good loss forecasting overtaken by a bad economy? However good the loss forecasting techniques are for projecting losses associated with a portfolio of booked accounts, inaccuracy will result when the quality of booked accounts is not understood. We must look at actual risk management practices to see whether accurate, informed, acquisition decisions are being made, to allow banks to manage their portfolios effectively.

Credit scores are used as a primary tool for deciding on whether to give someone credit, and may be used to help determine interest rate and amount of credit given. However, the question we need to ask is, do they cope well with a rapidly changing environment?

The traditional approach to risk management sets accept strategies according to the projected performance of each applicant, and interest rates reflect the risk (bad rate) associated with their credit score, resulting in losses like the example above.

Take a look Figure 1. It shows the bad rate on auto loans for each score range for three application points in time – June 05, 06 and 07. The bad rate is defined to be the percentage of auto loans that go to 90+ days past due within 18 months of opening. We observe a year-on-year increase in the bad rate for the same score cut-off strategy.

Despite the changing bad rate, the performance of the score, measured by its ability to separate out good payers from bad payers, changes very little. One commonly used scorecard had a change in the KS statistic of less than 2% across this period. The KS statistic is a measure of how well a score separates good from bad payers, or how well actual data fits to the model (click here or here for more resources on KS Tests). This shows the ability of a scorecard model to rank order risk is separate to its’ ability to predict bad rate at a given score.

If a scorecard continues to rank order well, why is a changing bad rate to score relationship a problem for lenders? It would not be a problem if they had perfect foresight on the bad rate associated with a score based decision being made when the decision is being made. Figure 2 shows the changing bad rate to score relationship across time, and was derived after time had passed and we could observe what the resulting bad rate actually was. Wrong decisions are made as a result of not having an accurate forward looking view.

When decisions were being made in June 2007, the bad rate lenders were expecting was the lower bad rate given in the blue line of the chart. However, those loans actually had the much higher bad rate associated with the green line. The consequences of this inaccurate information can be major. The consequences of an unexpectedly low bad rate may not always be as dire, but pricing will have been wrong, and opportunity for profitable business will have been lost.

In part two of this series we will discuss how the use of macroeconomic data can enhance the accuracy of bad payment rate forecasts. If you are interested in a monthly e-mails about blog updates, sign up for our newsletter. Click here to find out more about Equifax's advanced analytic insights.

This post was contributed by:

.
Vice President Modeling and Analysis

Churchill on Master Data Management

Sterling Metz — Thu, 20 Jan 2011 11:42:00 +0000

The precise Winston Churchill quote is, “To improve is to change; to be perfect is to change often.” The relevance is that accommodating change is the heart of reference-based keying systems.

End consumers are constantly changing and the various identification elements also are constantly in flux – address changes, name changes, and household composition changes, for example. How does the modern retail bank keep up with these changes? Fortunately, reference-based keying systems have a mechanism for doing just that.

First, any good reference-based keying solution is constantly adding newer identity data, since it helps improve its ability to relate all the various identification elements to a person. These additions often will lead to new associations and connections between identities. Second, a reference-based solution should have a mechanism for delivering these changes to you. Third, in order for your customer database to stay up-to-date, you will have to have a process for accepting changes to identities.

Here is an example: you have a prospect named Chip Consumer who lives at 123 Main Street. The reference database has Jon Consumer as 123 Main Street (Key #123456), but does not find Chip Consumer. Therefore, it will generate a new key for Chip Consumer (#777777) and save it separately. After all, this could be a brand new identity or it could be a variant of an already keyed identity – there is no way to know yet. Therefore, in your prospect database you use key #777777. A couple of weeks pass and the reference database gets some new information - Chip Consumer is really Jon Consumer (perhaps Chip is a childhood nickname). It will now add Chip Consumer as a variant of key #123456 since it is the same person. This is a brand new relationship. The reference database has to send some sort of update to you so you can benefit from this new information. It does this with a “was-is” record. In this example the update would say “the key that was #777777 is now #123456”. In this way, you can interact with Chip Consumer knowing that he is actually Jon Consumer.

A good reference-based keying solution will always try to improve itself by adding new data. It should have a clear and easy mechanism for tracking, recording and delivering new relationships to you so that you can benefit and your systems can take the next best action for the relationship. In this way, your view of your customers will be in a state of constant improvement.

If you are interested in a monthly e-mails about blog updates, sign up for our newsletter

This post was contributed by:

.

Mining the Gold in your Customer Base

Karthik Mani — Wed, 12 Jan 2011 21:01:00 +0000

This concept of cross-selling is neither new nor rocket science, but in this extreme business climate - more than ever – it is the lowest risk route to growth and higher profitability. That is, if you can do it well.

“The cost of us selling a product to an existing customer is only about 10% of selling the same product to a new customer.” - Dick Kovacevich, former CEO of Wells Fargo

US adults overall have 8.2 financial services products. On average they have 2.5 products with any single provider. Even the best banks service only 3.4 products per customer. Only 1 in 4 respondents would be uninterested in keeping all of the financial products and accounts with one firm. - North American Technographics benchmark Q1 2009

Mr. Moynihan's strongest attempt to present a new vision for the bank revolves around more cross-selling to customers. – Wall Street Journal quote of Brian Moynihan, new CEO of Bank of America

Sounds drop dead easy. You know – or should know - existing customers better than a new customer. Thus it follows that an existing customer is a lower marketing risk than using outside data to seek new customers. With no acquisition cost, you have a lower cost to market to them. You also should be able to provide a more-targeted offer than the competition. As the study referenced above shows, there is a high propensity for them to accept.

The reason many retail banks – as evidenced by the statistics above – fail to capitalize on this obvious strategy is that understanding your customers well enough to effectively cross-sell to them requires actions and capabilities often very difficult for many financial institutions.

Specifically, one must be able to view all the data about each customer in one place. This is essential to generating the understanding necessary for developing marketing strategies at a segment level.

A thorough understanding of current offerings and their success – or lack thereof - leads to creating the next, best action based on the individual customer’s propensity to buy and their response to the new model.

Additionally, achieving better relationship-based pricing similarly requires a thorough consideration of all the variables, from competition to your own value and price equations. Arriving at the best price through all these constraints and objectives also requires optimization techniques.

This brings us back to the essential necessity for technology that lets you consider all your data on each customer in one place.

For example, a retail bank that Equifax works with wanted to find additional methods of identifying profitable customers to supplement traditional credit data-centric marketing programs. They were not effectively leveraging their marketing data warehouse investment. They needed to expand beyond credit card to include other products. They could not readily identify consumers who also own a small business or have a demand deposit account in order to expand cross-sell capabilities.

Working with us, the company was able to key all customer and small business data in a manner that gave them access - in one place - to everything relevant to any given customer. They were able to implement a portfolio-analysis capability that helped them automate treatment plans, including the ability to make decisions around credit line increases, decreases and freezes. They were able to segment the population, identify credit-based “hot lists” and prioritize the contact queue.

All of these helped them get a reliable 360 degree view of the customer, giving them an understanding of the entire relationship for the first time. This enhanced their risk-management analytics and resulted in successful cross-selling of new products across all channels.

If you want to talk to an Equifax specialist about mining the gold in your customer base, click here to send us an email. If you would like future updates automatically sent to you, please sign up for our monthly newsletter. It summarizes the new articles in our blog.

This post was contributed by:

.

Karthik Mani brings 17 years of experience in Banking, Retail, and Manufacturing across analytics, optimization, and decision management. Currently he is responsible for strategy, new product innovation, and marketing for the technology and analytics solutions at Equifax.

Garbage In Is Still Garbage Out

Sterling Metz — Thu, 16 Dec 2010 00:01:00 +0000

In our previous blog entry, we discussed reference-based consumer master data management and how it can help you get a clear view of your retail banking customer despite name and address variances. As a result, you might be asking, “Why can’t I just try matching on my own? After all, I know that Jon is short for Jonathan. I know that Jane Doe at 123 Main Street and Jane O’Reilly at 123 Main Street is probably the same person and the difference in last name is because of marriage/divorce.”

Are you sure?

Maybe Jane O’Reilly just purchased the house at 123 Main Street from Jane Doe. Maybe Jon Doe is the son of Jonathan Doe. These nuances can drastically affect your interaction with your customer – whether that is marketing a new DDA solution, customer support, or credit/risk decisions. How can you make the next best action at the right time if you don't know who you are talking to?

That is why the source of reference-based keying is vital. The master-identity list or person-master file needs to be sourced with data that is both accurate and current. Because of those two requirements, we find a product based on credit-file data is the most reliable. A credit file is one of the most accurate sources for identity information available, and it is routinely updated so that changes (such as marriage and divorce) can be accurately included.

Getting a 360 degree view of your customer is important, but only if that view is correct. If you are having problems with disparate databases and multiple customer identities, using reference-based consumer keying based on credit data is the most accurate and current way to bring that view into focus.

How do you tie your customer data together?

If you are interested in a monthly e-mails about blog updates, sign up for our newsletter

This post was contributed by:

.

What Do Online Retailers Have That You Don’t?

Sterling Metz — Thu, 09 Dec 2010 00:01:00 +0000

When it comes to maximizing each interaction with a customer, few top online retailers.

From the initial splash page, to the individual item pages, to the check out page, online retailers offer suggestions of similar and complimentary products. The most noteworthy aspect is that each suggestion is tailored specifically to the customer. Every customer gets the next best offer given their buying history and the behind-the-scenes advanced analytics.

So, how can your demand deposit strategy cross-sell and up-sell as well as the giants of their industry?

In truth, they have an unfair advantage that retail banks don’t have – since online only retailers can leave tracking cookies to help them identify and link customers. Being able to confidently identify a customer is the foundation for superior next best action interaction models, but in brick and mortar businesses you can’t leave a tracking cookie with a real person. Or can you?

Every time a customer interacts with a teller, an ATM, or a branch manager, customer data is collected (including name and address). However, during each interaction, the customer data may be different. The customer may use Jonathan Consumer on their mortgage, Jon Consumer on their bank credit line, J.P Consumer for their DDA account, or maybe use a middle name as a first name such as Philip Consumer during another. The same is true for addresses that change over time.
Reference-based consumer keying can help with this problem. In a nutshell, a third party collects various name and address variations over time and uses that information to create a master identity reference database for you. Each unique identity in that database is assigned a unique key. To further the earlier example, Jonathan Consumer, Jon Consumer, J.P. Consumer and Philip Consumer would all have the same key since they are the same person. Using this key, a retail bank can link what would appear to be four different consumers and understand that they are in fact the same consumer. Only after you have a full view of your customer can you confidently and effectively cross-sell and up-sell through next best action technology.

Reference-based keying solutions can assign unique individual, household and address keys to every consumer, helping your company accurately identify customers regardless of address changes, name changes (married versus maiden), or name variations.

How have you leveraged customer data to build better offer management?

If you are interested in a monthly e-mails about blog updates, sign up for our newsletter

This post was contributed by Daniel Jean.

Daniel Jean is an AVP of Product for Equifax's Connexus product, a proprietary keying and master data management solution. Daniel has been in product management and product strategy for 12 years across the telecommunications vertical. Daniel hold an MBA from Notre Dame's Mendoza College of Business.

How a Business Rules Management System Can Help You Grow Profitable Customer Relationships

Andrew Skillen — Tue, 07 Dec 2010 00:01:00 +0000

A high priority for financial institutions is to identify their most profitable customer relationships, and grow and retain that customer base by taking the next best action for the customer experience. The core deposit relationship is more important than ever. Legacy technology platforms generally do not allow lenders to consolidate an ROI view of the consumer across all banking relationship. A first step is to gain a 360 degree view of consumers by integrating and keying across all behavioral, credit and transactional data. Then a business rule system can apply centralized credit risk policies against this more holistic view of the consumer.

Business rules management systems are emerging as a great way to mirror an organization’s credit policies and cross selling strategy, providing more control over how customer segments are treated. A business rules management system enables business users to respond quickly to changing market conditions using centralized policy management that offers flexibility over applying new customer credit policy treatments. Decision trees can automate customer segmentation and the application of new treatments in a way that mirrors your organizations existing credit policies and cross sell strategies. In short, the investment in your core processing system can be extended using a BRMS as a way to deploy more complex customer management programs across your retail bank.

For more information on how you can improve your customer management systems and make the next best offer for your customers, please check out our white paper on “Managing Risk in the New Economy”. You can also register for the upcoming Equifax/IBM webinar on using analytics and technology to gain deeper insight around maximizing customer lifetime value:
https://www.techwebonlineevents.com/ars/eventregistration.do?mode=eventreg&F=1002702&K=CAA1BC

Top 10 from "Retailution" - Highlights from BAI Retail Delivery Conference 2010

Sterling Metz — Tue, 09 Nov 2010 00:01:00 +0000

KCK Blog was in Las Vegas October 19th to October 21st, 2010 for the BAI Retail Delivery conference. Attendance was up from last year. The exhibit floor and sessions were packed. Overall, in public, there was a lot of talk about innovation, mobile banking, remote deposit capture, and social media. In one-on-one conversations, the focus remains on bread and butter items: how to sell better at the point of sales, how to replace the revenues lost due to the new regulations, ID fraud, and privacy. Here are the top ten things we heard at BAI.

10. Learning from other industries: Customer engagement is critical and there is a lot to learn from Zappos. What Zappos has achieved defines what is possible.

9. Try. Learn. Repeat.: According to Doug Merrill, former CIO at Google, Google has 200 experiments running in the search window at any one time. Most are too small to notice, but important for learning. How can we experiment, pull the plug fast when it doesn't work, and learn a lot?

8. Understand the latent need: Asking customers what they want is going to get what Henry Ford called the "faster horse" moment. Both Tom Kelley from IDEO and Doug Merrill talked about the need to learn from the customer but not asking them "what do you want?", but understanding latent need by looking with a fresh pair of eyes and observing what customers do rather than what they say are good ideas.

7. Work with customers to produce better results: Rilla Delorier, CMO of SunTrust Bank, talked about what helped them achieve the highest customer satisfaction numbers from their Reg E correspondence. They brought in some customers to help them draft correspondence to customers on overdraft protection.

6. The Red Queen Effect: Tom Kelley at IDEO talked about how in various industries the red queen effect from Lewis Carroll's "Through the Looking-Glass" is the apt metaphor for today - ""It takes all the running you can do, to keep in the same place."

5. Trust - depends on who you ask! - BAI President and CEO Debbie Bianucci said that 2/3rds of the US consumers have trust in their own banking institutions at BAI 2009. On the opening day of BAI 2000, USA Today published a survey in which the banks had 9% trust from consumers compared to the most trusted institution, the army which came in at 39%. I can't imagine the trust getting eroded by 57% in one year. I guess it depends on who you ask!

4. Data, Data Everywhere; but what should I do?: Everybody was talking about analytics but there was acknowledgement that little was being done. A number of sessions focused on leveraging that data to build a 360 degree view of the customer and cross-sell them more products to increase the wallet share leveraging next best action instead of next best offer.

3. On-line is the key battlefront for smaller banks: Credit cards have been consolidated by the large national players. 35% of the deposits are held by large national banks and that is increasing. The bank which has the highest name recognition in Louisville, KY is Bank of America. The kicker is that there is no Bank of America branch for 100 miles around! Smaller banks can't focus only on branch based banking in this environment and have to focus on online banking.

2. iPad Give Aways Work (or at least we vendors think): IPad giveaways were as popular as cold hard cash in the exhibit floor based on the number of vendors that were offering it and based on the number of attendees that were talking about it.

1. BAI can put up a show: BAI did a great show of bringing thought leaders from both within the industry and outside the industry that gave us a lot to think about. The video introduction to each of the general sessions was our personal highlight. Here is an example.

Of course, receiving the FINTECH100 award on behalf of Equifax as a top vendor to the Financial Services Industry was the most satisfying part of the whole trip. That said, there was a lot of learnings from the conference both from discussions with the attendees and from the sessions. Look for future articles here on those.

4 Reasons You Should Visit Equifax at BAI

Sterling Metz — Mon, 18 Oct 2010 00:01:00 +0000

For all attending BAI tomorrow, we encourage you to stop by and talk to Equifax. We are at booth 315 in the Mandalay Bay Conference Center. You can also learn about how we are integrating with IBM at booth MR4. Now for the four reasons:

4. Learn more about the demand deposit account solutions we have talked about on the blog
3. Pick up a printed copy of our white papers on effectively managing risk for retail bankers
2. Here firsthand about the Anakam acquisition and how to make multi-factor authentication a reality in your demand deposit strategy
1. Win a $250.00 American Express Gift Certificate

We look forward to seeing you there.

Technology Solution for Enterprise Account Management

Sterling Metz — Tue, 17 Aug 2010 11:17:00 +0000

As macroeconomic conditions have deteriorated, a retail banking risk manager’s job has become increasingly challenging. One of the more concerning changes in the credit industry is a significant increase in consumers that are walking away from their primary mortgage obligations. Technology helps lenders more effectively manage risk within their portfolio, in four key areas, which should be orchestrated together for maximum benefit:
1. Implement rapid change management.
2. Build a comprehensive view of the customer.
3. Leverage behavioral and credit data together.
4. Make use of advanced modeling techniques.

Follow the link to read the whole whitepaper, “Effectively Managing Risk in the New Economy”