Category Archives: FICO

Why FICO Was Named a Leader in Predictive Analytics

Great news, FICO was named as a leader in the March 2017 report, The Forrester Wave™: Predictive Analytics and Machine Learning Solutions, Q1 2017. This is an honor, and it underscores for us the value and positive business impact our analytic solutions continue to deliver.

Forrester Graphic Predictive Analytics Machine Learning Wave 300x240 Why FICO Was Named a Leader in Predictive Analytics

These types of technology evaluations are really useful for organizations needing guidance as they enter into new strategic ventures. With plenty of attention on the promise of advanced analytics and machine learning, this report from Forrester is very timely. The report highlights:

  • The Predictive Analytics and Machine Learning (PAML) Solutions market is “hot,” with Forrester forecasting “a 15% compound annual growth rate (CAGR) for the PAML market through 2021.”
  • The true essence and core value of predictive analytics, stating “enterprises that can make probabilistic predictions about customers, business processes, and operations will have an edge over enterprises that can’t.”
  • Machine learning is fundamental to artificial intelligence.

When you talk to data scientists and business leaders and hear their real-world stories, one begins to comprehend the possibilities that come with advanced analytics or data science. So, to compliment this Forrester Wave report, I thought I would share a few FICO case studies that demonstrate the value and competitive advantage predictive analytics can provide.

Airline Predicts and Solves Problems Before Crews Are Grounded and Customers Are Unhappy

Managing an airline that serves 100 million customers annually – with more than 3,900 daily departures during peak travel season to more than 100 destinations globally – is no small undertaking. And when regular operations become irregular, the task of keeping customers happy, flights on time and crews supported is daunting.

This is why airlines, like Southwest, have extensive contingency plans for irregular operations. Southwest has The Baker, a recovery optimization engine that turned the best practices of its irregular operations (IROPS) team into complex algorithms that solve potential challenges like maintenance problems or volatile weather, while minimizing the impact to passengers and crews.

Southwest can not only react to problems within minutes, but it can also get ahead of potential disruptions hours in advance and evaluate multiple scenarios. The Baker has helped Southwest improve on-time performance and customer satisfaction, while lowering flight diversions, tarmac delays and crew changes.

This is predictive analytics in action, and it has made a significant impact at Southwest, helping the airline gain advantage in a highly aggressive and competitive market.

Meaningful, Personalized Connections with Customers at the Grocery Store

Virtually any customer engagement can be improved with the power to predict. In the case of a major grocery retailer FICO works with, using analytics to deeply understand customers helped the retailer provide a highly personalized experience that encouraged loyalty and increased profitability.

The grocery retailer wanted to predict the propensity of a customer to purchase particular products and then prescribe the best offer. This required a multidimensional understanding of each of its 9 million loyalty program members, as well as taking into consideration business objectives and constraints.

With FICO’s advanced predictive analytics, the grocer was able to take its 380 billion possible offer combinations and deliver 20 tailored, relevant offers to each member every week.

Predictive Analytics Revolutionize the Energy Industry

Another great FICO customer example is SolarCity, a leading solar energy provider working to create an ideal way to store solar energy and then use it when demand is high. Electric demand can vary widely, based on things like changes in the weather, inconsistent supply or seasonal consumer patterns. With advanced predictive analytics, SolarCity is able to track and forecast complex scenarios for selling aggregated energy.

It’s fascinating that the same technology that is helping Southwest minimize flight delays and a grocery retailer deliver highly personalized offers to customers is also helping optimize a solar energy grid. Hopefully these real-world examples add a little perspective as you read the March 2017 Forrester Wave™ for Predictive Analytics and Machine Learning Solutions.

Let’s block ads! (Why?)


Do ‘Digital-Only Banks’ Have a Future in Asia?

digitalonly banking Do ‘Digital Only Banks’ Have a Future in Asia?

Imagine a scenario where banks offer their services digitally; not as an ad hoc feature but as a fully integrated mobile experience. A digital-only bank that allows customers to do everything on their smartphones, from opening a new account to making payments, settling credit card bills to resolving disputes, all without having to go to a physical branch.

If this vision sounds premature, then perhaps it’s time to update your view of what’s happening worldwide. This was the bold proposition offered by McKinsey & Company’s Sonia Barquin who presented at FICO’s Asia Pacific Chief Risk Officer Forum held this week in Thailand.

Banks worldwide are fighting back against fintech start-ups looking to cut their lunch with low-cost banking offerings. Barquin pointed out that some global banks are developing digital-only offshoots while others are bringing the philosophy and business model to their main bank offering.

Sonia encouraged the participants to be imaginative –and why not? Especially when she says her research shows that more than 80 percent of banking customers in Asia’s developed markets are willing to move part of their holdings to banks with a solid digital-only banking proposition?

While the financial benefits of the model for banks are obvious, the big question is, how do banks approach this massive undertaking? And especially the more traditional banks who have only just started to integrate many aspects of digital into their operations and offerings?

FICO’s decisioning technologies are clearly an advantage for banks considering an all-digital play. FICO has experience helping banks to quickly build digital offering which transform their business. Our experts at the CRO forum openly shared some great insights on how advanced technology can be leveraged in multiple areas of digital banking:

  • Instant credit decisions: Consumers expect an ‘instant’ experience on their mobile phones. FICO has been pioneering credit scoring for more than 60 years and now has the technology to assess the vast majority of consumers with some basic credit information. This empowers banks to issue credit cards, personal loans and other credit in the mobile channel with confidence. FICO is also working on solutions for those outside the traditional banking system, who may have a smartphone. As part of its financial inclusion efforts it is employing a range of partnerships and technologies to grant credit based on alternative data.
  • Pricing and loan optimization: FICO has been perfecting analytics that enabled banks to identify that sweet spot or optimal pricing for deposits, mortgages, car loans and other products. Finding the right balance between customer satisfaction and profitability needs to happen quickly when applications are happening online. FICO’s optimization algorithms and knowhow allow banks to make a variety of offers based on the customer’s risk profile, long-term value, loan term preferences and more.
  • Detecting fraud in real-time: Less exposure to fraud and financial crime when you are identifying customers online and not in person is critical. FICO’s real-time predictive analytics has been used for more than 20 years to help catch credit card fraud. This same technology is now being incorporated into FICO’s cybersecurity and financial crime solutions, so banks can spot out-of-pattern and suspicious behavior.

Ultimately, the digital-only banking model will not only require a behavioral shift among Asian consumers, but the opportunity for banks will require the deep integration of data and analytics into banking operations. The key to eliminating risk in the business model is to enable banks to make consistent, data-backed, and analytically powered decisions, and ultimately make the smartest choices. Once that happens, the digital-only proposition for banks could be a real game-changer in Asia.

Let’s block ads! (Why?)


Why the Panama Papers Leak Was Good for KYC

Panama Papers Why the Panama Papers Leak Was Good for KYC

For many banks, KYC — Know Your Customer — means asking them how they intend to use a product, where the funds are coming from for their new account, etc. At the same time, the bank will check the customer against sanction lists, PEP (politically exposed persons) lists, and so forth.

It’s not enough.

Some hard evidence that it’s not enough came in 2016, when the so-called Panama Papers leak revealed that thousands of people worldwide owned a shell company in one of the countries covered. This was, needless to say, not something those individuals had disclosed to their banks.

Should banks care? Absolutely. Under the KYC requirements that are part of current regulations, such as the 4th EU Money Laundering Directive and the fifth pillar of the BSA, the bank needs to know the business of their customers. If a customer owns an offshore company, it’s quite possibly so that they can avoid taxes — or, even worse, so they can hide the flow of money, potentially illegal money.

While there are some legitimate reasons for offshore companies, an offshore company by its very nature poses a risk that needs to be investigated. But you can’t investigate something you don’t know about.

Banks that don’t know about their customers’ offshore companies face not only non-compliance, but reputational risk. No one wants to read that their bank is supporting tax avoidance or money laundering by the use of a shell company in an offshore.

How do you check?

Immediately after the leak went public, banks began a hectic operation to find and fix. I know banks that hired teams of expensive external consultants to manually check if any of millions of customers owned a company in Panama.

A manual check has multiple disadvantages:

  • It’s costly and slow.
  • It might not stand up to an audit — who can reproduce what has been checked?
  • Worst of all, it’s a one-time approach. It will need to be repeated regularly, and intensively whenever further leaks are published.

We have a better way. We built an interface between the FICO® TONBELLER® Siron®KYC solution and the ICIJ Offshore Leaks Database. Not only is this database constantly updated, it’s exhaustive:

  • The data covers nearly 40 years.
  • It links to people and companies in more than 200 countries and territories.
  • It reveals more than 370,000 names of people and companies behind secret offshore structures.

Here’s how this works in the KYC process. During onboarding we automatically check if the potential client owns a company in an offshore country. If yes, we would raise a red flag and allow the financial institution to decide if they want to accept the customer. If accepted, the customer would be put in the high-risk segment, with enhanced due diligence. The customer would be monitored against very strict scenarios and thresholds.

If a customer doesn’t own an offshore company when they are onboarded but opens one later, our solution’s continuous customer screening against the ICIJ Offshore Leaks Database will automatically detect it and handle the case, following the risk appetite of the financial institution.

People are sneaky — sometimes a customer will open an offshore company and spell their name slightly differently to avoid the high-risk classification. We have an algorithm to catch that.

The Panama Papers and Bahamas Leaks are only the first step. Regulators will be demanding increased transparency into each customer’s business soon. That’s why we set up a future-proof integration — so that our clients can protect their reputations with an audit-secure, low-friction, scalable process.

The Panama Papers leak has strengthened KYC. We made finding out easy.

Let’s block ads! (Why?)


Hackers vs. Dracula: Biometrics Are No Silver Bullet

Dracula Hackers vs. Dracula: Biometrics Are No Silver Bullet

I’m not a big fan of vampire movies—I’d pick Blackhat over Abraham Lincoln: Vampire Hunter any night of the week—but there are a lot of similarities between hackers and vampires.

  • First, they’re afraid of the light. What hacker wants his true identity to be revealed?
  • Second, they suck the blood out of their victims. Whether stealing data or demanding payment for ransomware, “bloodsucking” is one of the kinder adjectives used to describe cyber criminals.

However, even though vampires are theoretically immortal, vanquishing them is pretty straightforward; any True Blood fan can tell you that a wooden dagger or silver bullet will do the trick. It’s not quite so easy to stop hackers in their tracks.

 Encryption can be effective …

 … but it’s not a stake through the heart of hacking.

Data encryption is a highly effective defense against hackers, particularly in achieving HIPAA compliance to protect Protected Health Information (PHI). By using sophisticated techniques such as (encryption) key rotation and key lifecycle management, hackers that manage to break in and get their hands on data can’t do too much with it because the data will be encrypted.

That’s true, but there’s one problem. Data needs to be decrypted to be used, and that means it’s easy for employee negligence to open the door to hackers. Unencrypted data is replicated across systems, stored in PDFs, saved on laptops and carried around on thumb drives. In searching for weaknesses in healthcare organizations’ storage of PHI, hackers first look for data that’s not encrypted, stealing it unnoticed in a crime of opportunity. Downstream, PHI fuels financial fraud by allowing cyber thieves to create false identities.

In 2015, the year in which an astounding 112 million patient records were stolen, employee negligence was the number-one security issue.

Biometrics are trending …

 … but they aren’t the silver bullet. (I realize some purists say silver bullets kill werewolves, not vampires, but stick with me anyway.)

To protect against consumer financial fraud, there’s a lot of buzz now about using biometric information — fingerprints, iris and facial recognition, and other unique physical characteristics — to authenticate payment card transactions. Here’s a typical cheer:

“Retail leaders should implement biometric authentication as an alternative to the EMV and other bankcards. Identity should be tied to a person — not a card. This is especially true in today’s omnichannel world where an EMV chip won’t protect fraud that occurs outside of a brick-and-mortar establishment.”

Like encryption, however, biometrics are not a silver bullet to stop hackers. As a defense mechanism, biometric authentication is actually worse because it can create a false sense of security. But once that information is corrupted or stolen by hackers, how do you prove who you really are? This excellent article in Scientific American captures the high-level privacy and cybersecurity implications that should be central to any discussion of biometrics:

“… [O]nce your face, iris or DNA profile becomes a digital file, that file will be difficult to protect. As the recent NSA revelations have made clear, the boundary between commercial and government data is porous at best. Biometric identifiers could also be stolen. It’s easy to replace a swiped credit card, but good luck changing the patterns on your iris.”

The best defense is vigilance

When it comes to ever-more resourceful and clever hackers, there is no single technology that can stop these criminals in their tracks. The best defense is constant vigilance.

All defenses can be compromised, and when that happens, you need to know it – and quickly. FICO® Cybersecurity Solutions deliver exactly that, allowing enterprises to identify emerging cyber threats and fight cyber crime in real-time. In doing so, FICO solutions help to reduce the dwell time of cyber threats, to dramatically narrow the window for potential damage.

To keep up with my latest musings on everything from vampires to hackers, follow me on Twitter @dougoclare.

Let’s block ads! (Why?)


Rise of the (IoT) Machines: When Do We Say “Enough”?

Terminator Rise of the (IoT) Machines: When Do We Say “Enough”?

Make no mistake about it, I am a technology-loving kind of guy. My job is all about helping people to make better business decisions using analytics technology, and I have boundless enthusiasm for speaking with FICO customers, data scientists and students about the power of behavioral analytics, artificial intelligence and machine learning.

But when it comes to my personal life, I’m a bit of a contrarian. There’s been a lot of news about studies showing how too much smartphone use, or too much time on social media, can have a stressing, depressing effect. So I purposefully limit my phone use and I’m not even on Facebook, Instagram, Snapchat, et al. (Although I do love using Twitter as part of my work, to connect with followers all over the world!)

How technology can rob joy

As the Internet of Things (IoT) matures, I feel like a similar trajectory is starting to take shape: initial euphoria over new technology — like Teslas in ludicrous mode, or being able to order Doritos for delivery via drone — that slowly turns to dislike and can eventually result in technology abandonment.

Here’s an example. In June 2015, WIRED magazine was absolutely agog over the June Intelligent Oven, “a $ 1,495 connected appliance that will quite literally cook your food for you.” WIRED is pretty effusive about June: “Using what amounts to a fancy meat thermometer, along with scales in each of the June’s four feet and an internal camera, the oven can figure out what you’re cooking, how hot it is, and how much it weighs. And if you know those three things, you can cook almost anything.… Eventually, the team hopes, you’ll just pop something into the oven, tap to confirm it is what the camera thinks, and walk away until a push notification tells you it’s almost ready.”

The article then asks, rhetorically, “Doesn’t this kill the fun, the experimentation, the trial-and-error that makes cooking great?” To which, “[t]he June team says no, that the idea is to that job so you can focus on the other stuff. And… let’s be honest: This product is mostly for people who wouldn’t otherwise cook much. It’s telling that the team’s two demos are a plain bagel and six cookies; the goal, at least immediately, is more ‘don’t burn the toast’ than ‘award-winning roast.’”

OK, I get it. You could call it “an oven for dummies,” to appropriate an analog from the insanely popular book series. But what’s wrong here, to my mind, is that by thinking or doing too much for “dummy” us, IoT machines can rob us humans from one of life’s greatest joys: learning something new.

(A few months later, FastCompany took umbrage with the June oven too, saying, “This $ 1,500 Toaster Oven is Everything That’s Wrong With Silicon Valley Design – Automated yet distracting. Boastful yet mediocre. Confident yet wrong.” Ouch.)

Failure can be delicious

Failure is, and should be, an expected and accepted part of life. For instance, cooking: With the possible exception of Martha Stewart, anyone’s first time baking cookies, or making any recipe, is bound to have some kind of surprise, often unpleasant. What kid hasn’t made that Toll House cookie recipe on the back of Nestlé’s chocolate chips, and had the cookies come out unusually crispy, and black on the bottom?

Yes, those cookies looked awful, but they still tasted good. And the next few times you made that recipe, you probably experimented with using more flour or lowering the oven temp. Eventually you got the recipe and technique right, and now take pride in your ability to make a delicious chocolate chip cookie from scratch. The people who eat your cookies no doubt shower you with praise, and you bask in it. What’s not to love?

Learning gives us satisfaction

In my experience, learning makes people happy. It’s wonderful when we can do this at our jobs, and fulfilling when we can learn in unexpected ways. Ceding too much responsibility to IoT devices robs us of everyday learning opportunities.

One of the learning experiences that gives me much satisfaction is racing my car, a 2003 Porsche 911 Targa, model 996. They call it a “widowmaker”: no stability control, no traction, nada. The Targa is a challenging car to drive at high speeds on an autocross course, and learning how to handle it was not easy. But I’ve kept at it over the years, and each time I hit the track I drive a little faster, and with a little more confidence. It gives me great satisfaction to know I’ve mastered the car, instead of letting a car just drive itself. (Porsche’s CEO agrees.)

Come hang out with me at Willow Springs on April 28, when I’ll be racing my widowmaker. And follow me on Twitter @ScottZoldi.

Let’s block ads! (Why?)


Are Your Fraud Alert Replies Being Spoofed?

Spoof SMS reply Are Your Fraud Alert Replies Being Spoofed?

As banks try to improve customer experience, reduce fraud and cut operational costs through interactive SMS, criminals have moved in to take advantage of the channel. The latest fraud scam involves ‘spoofing’ CLI (calling line identity) numbers to respond to SMS fraud alerts intended for customers.

“Spoofing” SMS or texts might seem like something teenagers would do, perhaps sending fake texts on Valentine’s Day appearing to be from someone else. Instead, what’s happening is more sinister.

If a credit/debit card transaction is deemed as suspicious, banks can alert customers through SMS, as well as through automated voice, mobile application push notifications and emails. If the transaction is genuine, the customer simply needs to respond to the SMS to confirm this, without actually having to speak to an operator in a call centre.

What the fraudsters are doing is making a fraudulent transaction using a compromised card and then successfully ‘spoofing’ a customer’s SMS response, confirming the transaction to be genuine when it isn’t. The fraudsters don’t know for certain that the customer got an SMS alert in the first place – but they might know the bank’s alert and customer notification strategy. They would have to have obtained the customer’s telephone number on the black market, possibly when they would have obtained the credit/debit card details. The fraudster then guesses the correct timescale in which to ‘spoof’ the response, before the genuine customer can reply.

FICO are fully aware of this emerging fraud threat and have a range of solutions available as part of our FICO Fraud Resolution Manager:

  • Our SIM Swap solution detects whether the SIM card may have been ‘hijacked’ by a fraudster
  • SMS carousels consist of a range of rotating numbers which prohibit the fraudster from ‘spoofing’ one known number
  • A PIN/OTP (one-time password) request provides reassurance that the alert has reached the right person

Contact us if you would like more information on any of these redresses.

Let’s block ads! (Why?)


The Skinny on Trump’s Regulatory Reset

Regulatory Reset Image blog The Skinny on Trump’s Regulatory Reset

In my 2017 regulatory predictions post last month, I concluded by saying that the new year would be very different for the financial services industry than 2016. This certainly didn’t take long to come to fruition. In the first two weeks of the new administration, President Trump took several steps aimed at slowing down as well as scaling back current and future regulations. Despite these aggressive actions, there remains a number of challenges related to the reach and impact of these directives.

Regulatory Reform through Memorandum and Executive Orders

Out of the gate, the Trump administration made good on its promise to curtail the pace of federal regulations. Assistant to the President and Chief of Staff Reince Priebus issued a memo on Inauguration Day that, in part, calls for the heads of executive departments and agencies to initiate a regulatory freeze until someone designated by the President has a chance to review and approve them.

The freeze was followed by an Executive Order (“1 in, 2 out rule”) issued 10 days later that required for each new regulation issued by an agency, at least two prior regulations be identified for elimination. The executive order further mandates that the financial impact of each new regulation be offset by the savings from the rescinded regulations, such that the total incremental cost not exceed zero.

Finally, on February 3, the President signed an Executive Order that details six “core principles for regulating the United States financial system.” The Executive Order directs the Secretary of the Treasury to consult with the heads of Financial Stability Oversight Council (e.g., OCC, FDIC, Federal Reserve, CFPB, FHFA and SEC) and report to the President within 120 days the extent to which current laws, regulations and oversight requirements, including those connected with the Dodd-Frank Act, help promote the six core principles.

Application Limited to Executive Branch Agencies

Since the term “department or agency” is not defined in either the memo or the first Executive Order, there was some initial confusion as to whether the requirements of these presidential directives extended only to executive agencies, or also to independent agencies such as the CFPB, OCC, FDIC, SEC and Federal Reserve. A few days after issuing the “1 in, 2 out rule,” the White House issued interim guidance explicitly stating that the order’s provisions did not apply to independent agencies. However, this was not before the Federal Reserve had challenged the reach of the regulatory freeze and the “1 in, 2 out rule” on January 30 by announcing a final rule that no longer subjected select regional banks to the quantitative portion of the annual Comprehensive Capital Analysis and Review (CCAR) stress test.

Existing Challenges and Potential Impact

Despite the swift moves taken by President Trump, it is not clear how much impact these actions themselves will have on slowing down and reducing the number of regulations. While a regulatory freeze will temporarily halt the rulemaking progress at all executive branch agencies, as previously stated, the freeze does not extend to independent agencies.

The “1 in, 2 out rule” has been critiqued as being complicated and difficult to implement. For example, many regulations are mandated by statute and cannot be repealed without an act of Congress.  As a result, this may significantly narrow the pool of existing federal regulations that are eligible to be rescinded. Also, any repeal of an existing regulation must go through the extensive and deliberative process required by the Administrative Procedure Act. Already, this Executive Order is facing a legal challenge.

In addition, while the Treasury Secretary can make determinations of what regulations align with the six core principles detailed in the February 3 Executive Order, any changes to Dodd-Frank Act regulations will require congressional action. Though the Republicans control both chambers, Democrats have large enough numbers in the Senate to make it very difficult for Republicans to garner the 60 votes necessary today to get controversial legislation adopted. Not surprisingly, Republicans are actively discussing ways to clear this obstacle.

Where the Real Power Lies

The Trump administration’s largest impact on shaping the regulatory agenda in financial services may lie outside of its executive orders and memoranda. An obscure 1996 statute, the Congressional Review Act (CRA), has suddenly become a potentially powerful tool.

Under the CRA, before a new regulation takes effect, Congress can take action (through a simple majority vote) on a joint resolution disapproving the rule. If the joint resolution is approved in both chambers, the resolution is sent to the President for signature or veto. If the regulation is invalidated, the Agency that issued the regulation is prohibited from issuing a substantially similar regulation without a subsequent act of Congress.

Since its inception, only one time under the CRA had a resolution been signed into law by the President. During the first three weeks of the Trump administration, the House has passed 13 resolutions and two have already been signed into law.

It is easy to envision Republicans using the CRA to invalidate, for example, any newly issued CFPB regulation that they oppose. In addition, within the next 12 months, President Trump will be able to appoint new leaders at the OCC, Fed and FDIC. He has already nominated a new head of the SEC. New leadership can reshape the regulatory agenda of an agency, and this may be one of the most impactful tools that the administration has in ensuring that its regulatory reset is realized.

We are just a few weeks into a new administration, and already we have witnessed a dramatic shift in approach. Yet the action is just heating up as Congress prepares for the introduction of Dodd-Frank reform legislation, and President Trump continues to deliberate over the future of CFPB Director Richard Cordray. The outcome of these battles will help determine the expansiveness of Trump’s regulatory reset agenda.

Let’s block ads! (Why?)


Is Your Analytics Supply Chain Broken?

Analytics Supply Chain Is Your Analytics Supply Chain Broken?

This is a guest post from Thomas H. Davenport and FICO’s Zahir Balaporia. A version of this post was also published on Data Informed.

Businesses across many industries spend millions of dollars employing advanced analytics to manage and improve their supply chains. Businesses look to analytics to help with sourcing raw materials more efficiently, improving manufacturing productivity, optimizing inventory, minimizing distribution cost, and other related objectives.

But the results can be less than satisfactory. It often takes too long to source the data, build the models and deliver the analytics based solutions to the multitude of decision makers in an organization. Sometimes key steps in the process are omitted completely. In other words, the solution for improving the supply chain –  advanced analytics – suffers from the same problems that it aims to solve.

Therefore, reducing inefficiencies in the analytics supply chain should be a critical component of any analytics initiative in order to generate better outcomes. Because one of us (Zahir) spent 20years optimizing supply chains with analytics at transportation companies, the concept was naturally appealing.

More broadly speaking, the concept of the analytics supply chain is certainly applicable outside of its namesake business domain. It is agnostic to business and analytic domains. Advanced analytics for marketing offers, credit decisions, pricing decisions, or a multitude of other areas, could benefit from thinking about deploying advanced analytics using the supply chain metaphor.

Steps in the Analytics Supply Chain
Analytics can easily be viewed in supply chain terms. In the analytics supply chain, the customers are decision makers and the products being consumed are analytical models. The analytics engines that serve up recommendations or solutions are akin to manufacturing, turning data into consumable decisions. Data are the raw materials that enable us to generate the analytical models. The outputs of this supply chain are better decisions—ideally embedded into business processes and systems so that they can be performed repeatedly.

Let’s take this one step further, starting with data. We all know what happens to physical supply chains when we don’t have the right raw materials, especially of the right quality. When your data has low quality, so does your analytics – and, ultimately, the decisions you make. Beyond intrinsic quality, reliable access to raw materials is equally important. As product supply chains stretch across the oceans, data supply chains stretch across multiple systems and firewalls. Integration of such data can be time-consuming and expensive, and interruptions in the data supply chain can be very disruptive to decision making. And if decisions are sensitive to more real-time updates, then having data delivered with low latency is just as important as a key component arriving on time at the loading dock.

As we’ve mentioned, analytical models have much in common with manufacturing processes. These models are the machines that transform data into consumable predictions, recommendations and insights. The quality of those predictions, recommendations and insights relies on the same attributes as the quality of finished manufactured products. They must be fit for use, offered at a cost that attracts consumers, and made available when the consumer wants or needs them.

Analytics supply chains that take too long, or are too expensive to deliver the solutions that decision consumers need, get replaced with solutions that are more readily available. These solutions are often end-runs around the enterprise-level analytical infrastructure. These alternatives might range from “gut feel” to a spreadsheet. The spreadsheet is low-cost, and the quality seems fit for use because the higher-quality solution isn’t available, will take too long, or is too expensive. What many users of spreadsheets don’t realize is that they are prone to errors—between 20 and 80% of spreadsheets have been found to have errors in several research studies. And they lead to proliferation of different versions of the truth around an organization.

Finally, an analytics-driven decision is the finished product in the analytics supply chain, so let’s think about deployment of analytical solutions like product distribution. Just as products sitting in warehouses don’t deliver bottom-line results, models sitting on an operations researcher or data scientist’s computer that can’t be deployed efficiently will not deliver value to your decision making.

For example, the Netflix Prize engendered a model that improved the ability to predict user ratings films by over 10%, but it was never implemented because it was too complex. And as the number of models grows due to technologies such as machine learning, the ability to manage the growing inventory and distribution of models to support decision making will become more important in managing your analytics supply chain. Hiring more advanced analytics professionals would help, but that labor supply chain has its own limitations and should be viewed as an important constraint in designing an analytics supply chain.

Benefits of the Supply Chain Perspective

The primary benefit of considering analytics as a supply chain is a change in an organization’s perspectives and processes for doing analytics. It means that an organization can take a broad, holistic perspective on the use of analytics, and won’t develop “local optimum” capabilities that don’t benefit the entire process.  We’ve seen companies, for example, that hired super-smart data scientists who can “manufacture” many complex analytical models at a rapid rate. However, because of varied types of barriers (policy/technology/data/scalability/complexity), the company was unable to deploy those models.

Just as supply chain-focused companies measure the performance of their supply chains, those with an analytics supply chain perspective can measure the performance of their broad analytical processes. They can measure inputs (number of models created, number of analysts) as well as outputs (decisions affected by analytics, business value achieved from those decisions). They can rapidly identify bottlenecks and areas of under- and over-capacity. They can measure and improve the “time to insight, decision, and action” of the analytics supply chain for particular problems and decisions.

A large manufacturing company, for example, was in the process of deploying a distribution planning and optimization system. The prototype model was complete and the IT plan to deploy the system was estimated at 10 months and 4 FTEs. By leveraging a new analytics modeling and deployment platform, they were able to cut the deployment time and resource need by 50%, effectively a 75% savings from the original estimated cost.  Using this deployment capability has given this company a competitive edge in its analytics supply chain by cutting “time to decision” significantly.

Just as companies apply information technology to optimize and automate their supply chains, technology can also benefit the analytics supply chain. Machine learning, for example, is primarily a means of automating the production of analytical models. It can be a substantial aid to an analytical supply chain if the organization employing it is able to deploy the resulting models and embed them into business and decision processes. Companies that employ “model management” technology have realized that keeping track of the assets in the analytical supply chain is just as important as tracking inventory in the physical supply chain.

Is Your Analytics Supply Chain Broken?

How do you know if you have an analytics supply chain problem? The following points should serve as a simple initial diagnostic checklist.

  • If your ability to deploy advanced analytics is stifled because you need IT, data, and/or other scarce technical resources to respond to changes in your business environment, then you probably have an analytics supply chain problem.
  • If the time it takes to modify, calibrate or maintain advanced analytics systems is cutting deeper and deeper into your time to develop and deploy the next set of models, then you have an issue with the back end of the analytics supply chain.
  • And if you work in advanced analytics and have been preaching about optimizing everyone’s processes but your own, then it is time to focus on your analytics supply chain.

Moving the word “analytics” forward in the phrase “supply chain analytics” may seem trivial. But a small shift in the location of a word can make a big difference on where you focus your attention.

Tom Davenport, the author of several best-selling management books on analytics and big data, is the President’s Distinguished Professor of Information Technology and Management at Babson College, a Fellow of the MIT Initiative on the Digital Economy, co-founder of the International Institute for Analytics, and an independent senior adviser to Deloitte Analytics.

Zahir Balaporia is a Solutions Partner on the FICO Optimization team. Prior to joining FICO, he spent 20+ years designing and deploying analytics solutions in supply chain, transportation, and logistics with a focus on deployment and change management. He can be reached at

Let’s block ads! (Why?)


AI Meets AML: How the Analytics Work

The focus on financial crime, and the money laundering that funds terrorist attacks and other criminal activities, has forced the industry to look for smarter approaches. In the previous posts in this mini-series, TJ Horan noted that AI is the newest hope for compliance, and Frank Holzenthal explored the benefits that AI can bring to compliance officers.

Now it’s my turn, and I’m going to explore the AI and machine learning technologies my team has integrated into the FICO TONBELLER Anti-Financial Crime Solutions. We have built on top of the FICO TONBELLER solutions using FICO’s battle-proven and patented artificial intelligence and machine-learning algorithms, which are used in FICO Falcon Fraud Manager to protect about two-thirds of the world’s payment card transactions.

Industry experts have begun to realize the significance of analytics in combatting anti-money laundering. For instance, Aite Group LLC in its 2015 report Global AML Vendor Evaluation noted that “increasingly, regulators recognize that rules alone are not an effective manner of detection and are pressuring banks to include more sophisticated analytics.” Our new machine-learning techniques are directed specifically towards real-time, transaction-based KYC anomaly detection and highly refined self-learning models focused on anti-money laundering SAR (suspicious activity report) detection.

As Frank noted in his post, we have integrated two main AI components into our AML products.

Soft-Clustering Behavioral Misalignment Score: Powerful customer segmentation using Bayesian learning

Traditional AML solutions resort to hard segmentation of customers based on the KYC data or sequence of behavior patterns. FICO’s approach recognizes that customers are too complex to be assigned to hard segments and need to be monitored continuously for anomalous behavior.

Using a generative model based on an unsupervised Bayesian learning technique, we take customers’ banking transactions in aggregate and generate “archetypes” of customer behavior. Each customer is a mixture of these archetypes and in real time these archetypes are adjusted with financial and non-financial activity of customers.  We find that using clustering techniques based on the customer’s archetypes allows customer clusters to be formed within their KYC hard segmentation.

Different clusters have different risk, and customers that are not within any cluster are suspicious. As transaction monitoring is applied differently based on KYC, understanding which customers are misaligned within this segment based on their transaction behavior is very important, and transaction rules may not be sufficient. Further, as customers deviate from known clusters of typical behaviors based on transaction archetypes, the Soft Clustering Misalignment score alone can be used to drive investigations and possible SAR filings.

Clustering AI Meets AML: How the Analytics Work

AML Threat Score: Using machine learning to detect and rank-order suspected AML cases

Our AML Threat Score prioritizes investigation queues for SARs, leveraging behavioral analytics capabilities from Falcon Fraud Manager. It uses transaction profiling technology, customer behavior sorted lists (BList), and self-calibrating models that adapt to changing dynamics in the banking environment. Here’s how these work together:

  • Profiles efficiently summarize each customer’s banking transaction history into behavioral analytic features using recursive analytic algorithms, making ultra-low latency, real-time analytics possible.
  • BList maintains a weighted list of the most frequent accounts for a customer, based on regularity and frequency of money transfers to those accounts. Features constructed from BLists act as “fingerprints” of the customer’s account. Changes in BList can indicate suspicious behavior to be investigated. In a data consortium of multiple banks, BList can be used to create global “fingerprints” of good and bad accounts, allowing for quickly identifying fast-changing patterns of malicious activities that couldn’t otherwise be detected.
  • The streaming self-calibrating models are capable of automatically tracking the outlier points for each feature in real time. Then a Multi-Layered Self-Calibrating (MLSC) Score is created by combining multiple outlier models from all the features. MLSC models are structured like a neural network model, where the features in the hidden nodes are selected to minimize correlation. The weights of the model are either expert-driven or based on limited SAR data.

BList AI Meets AML: How the Analytics Work

These capabilities enable the AML Threat Score to analyze transactions in real time, pinpoint even previously unseen money-laundering patterns and stop them before the transactions are executed. The score can also be used to layer on top of current rules-based transaction monitoring to help with prioritization of investigation, to improve the detection and lower false positives.

Multiple ways for adopting analytics using Siron®

FICO’s real-time and dynamic KYC risk assessment and SARs-filing capabilities are unmatched in the market today. Our Siron® product also provides rules-based transaction monitoring capabilities that are relied on by both financial institutions and by regulators.

Siron allows for a gradual integration of analytics into current transaction monitoring. In a “rules first” approach, these novel techniques can be used to reprioritize the rules-based alerts for more effective workload management. A “scores first, rules second” approach can detect new types of AML patterns in real time while potentially reducing the rules set. The ultimate in this value proposition is the adaptive model framework, where FICO’s self-calibrating models are fully leveraged to learn from recent SARs while the rules respond to the regulatory typologies.

FICO has had a history of firsts, from being the first to use analytics for credit application processing systems in the 1950s to being the first to use machine learning for fraud detection in 1991. With our AML analytic capabilities, we continue to be the trailblazers!

Follow me on Twitter @ScottZoldi

To find out more about how FICO is putting AI to work in AML, register for our February 23 webinar, co-presented with CEB TowerGroup, “Hiding in Plain Sight: Is Your KYC Process a Spotlight or a Blindfold? Operational Benefits of New Analytic Technologies.”

Let’s block ads! (Why?)


Financial Health: The Key to the Future of Customer Acquisition

discoverscorecard Financial Health: The Key to the Future of Customer Acquisition

For all the differences between traditional banks and new online lenders, there is one business challenge that unites both groups: acquiring new customers.

Despite the advancements made by online lenders in many traditional financial services processes (such as account opening), the methods most online lenders use to advertise to and acquire new customers come straight out of the bank marketing playbook – third-party brokers, television ads, and even direct mail solicitations.

While online lenders experiment with offline marketing tactics, banks are aggressively ramping up their online marketing efforts, with total digital ad spend by U.S. financial institutions predicted to top $ 10 billion in 2019 (up from $ 5.3 billion in 2013). This shift is being motivated by continuing declines in branch activity and the increasingly-evident risks of branch-based sales strategies.

The result is that both groups are converging on a set of established acquisition techniques that are compatible with a post-branch environment. The problem, of course, is that this convergence will drive up all parties’ acquisition costs. While more intense competition in these marketing channels will be great for Google, Facebook, and the U.S. Postal Service, it will likely prove unsustainable for both banks and online lenders in the long run.

A new approach is needed. One that is aligned with the digital-centric preferences of today’s consumer, sustainable from a cost perspective, and able to differentiate your company from the numerous competitors jostling for your prospective customers’ attention.

Several leading banks and online lenders have started to experiment with a new approach by providing free financial health and planning tools to customers and non-customers alike. This approach enables these lenders to build a “user base” of engaged prospects who start to look to the lenders for information about their financial health and recommendations for improving it. This puts these lenders in the privileged position of knowing users’ financial situations and being able to (eventually) make relevant product and service recommendations that will help users meet their financial goals.

Here are a couple of interesting (though nascent) examples from both online lending and traditional banking:

  • Prosper Daily – Prosper Marketplace made headlines in September of 2015 with its acquisition of BillGuard, a personal finance app provider, for $ 30 million. The company’s mobile app (now branded as Prosper Daily) allows consumers to monitor their credit and debit card transactions for unauthorized charges or other signs of fraud or identity theft. The app, which had approximately 1.3 million registered users at the time of acquisition, gave Prosper a large base of engaged consumers to cross-sell its core loan products to. This acquisition opportunity is apparently promising enough for the company to continue fully supporting Prosper Daily, despite significant cut-backs in other areas due to declining loan volumes in recent quarters.
  • Discover Credit Scorecard – Building on its success offering FICO® Scores to current customers through the FICO Score Open Access program for free, Discover recently became the first credit card issuer to offer FICO® Scores to non-customers at no charge as well. The Discover Credit Scorecard enables Discover customers and non-customers alike to access their FICO® Scores and understand the key factors that are impacting their FICO® Score, for free, through an online portal. The extension of free credit scores to non-customers is becoming more common among leading credit card issuers because, despite the costs, it gives these lenders a valuable new avenue for engaging prospects around their financial health and goals.

If you use either of these services, it becomes apparent that Prosper and Discover are more focused on providing valuable financial management assistance to users than they are on marketing or cross-selling. This patience speaks well of both companies. Developing a critical mass of highly engaged users takes time and can be easily undercut by early attempts at monetization. This is particularly true in financial services, where users are looking for services that they can trust to prioritize their personal financial health and success over the short-term profits of the companies providing those services. Establishing that trust is critical because it opens the door to carefully crafted, highly targeted product and service recommendations down the road.

In a way, these new financial health tools are trying to establish a modern replacement for what bank branches used to provide – a trusted environment for financial services and advice.

Banks and online lenders looking to replicate this approach (and eventually monetize it) will need a couple of key ingredients:

  • A standalone personal financial management (PFM) tool, app, or service. There are a lot of PFM activities that you can help consumers with – monitoring credit scores, identifying fraudulent transactions, paying bills, establishing budgets, automatically sweeping money into savings accounts, etc. The key is to focus on a specific area rather than trying to be all things to all people. First-generation PFM tools failed to gain meaningful user adoption because their capabilities were too broad and overwhelming to use.
  • Sophisticated data management and analytic capabilities. In order to iteratively improve your PFM offering and mine insights that will drive targeted acquisition campaigns, you need to understand how consumers are using your tool, app, or service and what that usage tells you about their broader financial needs. The ability to accurately predict the type and timing of future product needs is obviously critical.
  • Real-time decisioning and offer management capabilities. When the time comes to monetize your PFM offering, you will need to be able to act on the analytic insights that you have been gathering on your users. Can you pre-approve users for specific products in real-time in order to respond to time-sensitive marketing triggers? Can you optimize the parameters of those product offers to maximize customer interest, satisfaction and profitability? Can you present the offers in a compelling and non-intrusive way?

When the question is how to develop a new, differentiated approach to customer acquisition, the truth is that there is no easy answer. However, at a time when banks and online lenders are increasingly competing for the same customers in the same marketing channels, we know that this question is of paramount importance.

While it is too early to know definitively what the future of customer acquisition will look like, the financial health initiatives at Prosper Marketplace and Discover give us an intriguing hint.

FICO is a Gold Sponsor at LendIt USA 2017, which will take place on March 6-7, 2017 in New York City.

Let’s block ads! (Why?)