I’m still shocked when I see how often operations use a week as a unit to determine the next time to take action on an account. And people wonder why they never get in touch with debtors.
[Read More]
In the world of customer centric targeted marketing, the biggest challenge that marketers face is that of balancing relevance with return on the investment (ROI). Relevance implies that the targeted message speaks to the targeted customer, both in terms of content as well as timing. For example, if a bank reaches out to a customer with a message about taking out a mortgage at an attractive interest rate, it would be relevant only if the customer needs the mortgage in the near future. An irrelevant message can turn off potential customers and can do more harm than good.[Read More]
The title of the book best summarized the lesson: Identify the meaningful signals and avoid the noise. This is precisely the problem that Big Data presents to analytic professionals (and non-professionals) today. In a world where you can collect everything, should you? Probably not.
[Read More]
For almost 30 years, I’ve visited collection and recovery operations all over the world to demonstrate software. I’m still shocked when I see how often operations use a week as a unit to determine the next time to take action on an account. The next steps in workflow are often set to 7, 14, or 21 days from the current date.
And people wonder why they never get in touch with debtors.
My mom plays bridge every Wednesday morning and goes grocery shopping every Wednesday afternoon. She’s never home on that day of the week. If you plan on collecting from her (and I hope it doesn’t come to that), you’ll need to call on a different day of the week.
This is Collections 101: Don’t use multiples of seven for callbacks if you want to recover debt!
It would be great if you could call a consumer every 6... [Read More]
In the world of customer centric targeted marketing, the biggest challenge that marketers face is that of balancing relevance with return on the investment (ROI). Relevance implies that the targeted message speaks to the targeted customer, both in terms of content as well as timing. For example, if a bank reaches out to a customer with a message about taking out a mortgage at an attractive interest rate, it would be relevant only if the customer needs the mortgage in the near future. An irrelevant message can turn off potential customers and can do more harm than good.
Predicting the Right Timing
This problem is a predictive problem where one not only needs to predict the likelihood that the customer would require a mortgage, but also get the timing right. Target the customer too early with a mortgage offer when she is not yet ready for the offer, and it’s... [Read More]
A few years ago, following his success in predicting the results of the 2008 election, poll analyst Nate Silver wrote a book called The Signal and the Noise. The book outlined his approach to understanding the limitations of polling to represent an elections ultimate outcome. By better understanding what and to whom pollsters ask questions, he devised a methodology to focus on polls that, in aggregate, best represented the voting electorate and, more importantly, avoid those polls that historically skewed their results to benefit particular constituencies.
The title of the book best summarized the lesson: Identify the meaningful signals and avoid the noise. This is precisely the problem that Big Data presents to analytic professionals (and non-professionals) today.
In a world where you can collect everything, should you? Probably not. Recently, ZDNet’s Stiligherrian wrote an article in which he discussed the Big Data “collect everything mentality” that grounds many Hadoop... [Read More]
I’ve been blogging about the growing role of analytics in strategic planning. In my last post, I shared a success story from a client that was able to sharpen its strategic vision in order to recognize and avoid a looming profit cliff. In this post, I’ll discuss our work with a bank that, after acquiring a portfolio, had to quickly counter performance trends that could reduce ROI on this asset.
This regional US bank—who I’ll call Bank B—had acquired a credit card portfolio with a prevalence of high-scoring customers, indicating low credit risk. With conversion requiring a lot of staff and analyst time, the bank asked FICO to help with a pre- and post-conversion strategic plan. Executives wanted to ensure there would be no negative surprises preventing the bank from hitting its ROI target for this new asset.
FICO consultants performed a cross-lifecycle analysis of portfolio profit components and forecasted... [Read More]
Our latest infographic identifies the eight characteristics of a Big Data Analytics Software (BDAS) developer. BDAS developers write the code that sits in every analytic application, service and solution, powers the analytic engines of the future, and turns massive complex data streams into clear insights in real-time. The characteristics are: RADical programmer (rapid application development that is), teamster (good in the scrum), parallel processor (high performance computing and parallel processing), captain of chaos (love for NoSQL), polygot (codes in multiple languages), open sorcerer (Hadoop and Hive is their domain), go Git-er (finds it on GitHub and BitBucket), and fearless explorer. Check out the infographic to see if you have what it takes to be a BDAS developer.
The Internet of Things was a hot topic at this year’s CES, a global consumer electronics and consumer technology tradeshow. The new devices at the event, included a Wi-Fi-connected ceiling fan controlled by a Nest Learning Thermostat, automated door locks, light switches and LED bulbs, smart sports apparel and the ilk.
The Internet of Things will produce an astounding mass of data, but there is no value in that data until data-processing technology and analytics are ready to take on the challenge in real time. In fact, the decision value of data is short-lived. The sheer volume, velocity and variety of Big Data make it extremely challenging to find the right data for your decisions, let alone act on that data while the decision is still relevant.
That’s where streaming analytics and real-time data processing come in. And it is what makes today’s announcement around the FICO® Data Management Integration... [Read More]
The world runs on analytics. Every business decision is increasingly made based on data, and every application, service and solution will soon use data analysis to respond to what’s happening now and anticipate what will happen in the future.
Yet, the pool of people capable of using and deploying Big Data analytics needs to expand beyond the “quants” – quantitative analysts, data scientists, analytic scientists and the like. Big Data Analytics Software (BDAS) needs to be accessible to all. Just as the desktop computer revolution made computers easy to use, and computer skills attainable for every worker in the ‘80s and ‘90s, the BDAS revolution will put analytics in the hands of every professional in the workplace of the future.
Last year IDC reported in its “Trends in Enterprise Hadoop Deployments” that only 32 percent of enterprises had actually deployed Hadoop, with another 36 percent planning to do so in... [Read More]
As UK card spend picks up in a healthier economy, is this balance growth healthy?
To investigate this issue, our team that manages the FICO Benchmark Reporting Service did some digging. The measure we looked at compares growth in delinquent balances vs. growth in current balances. The higher this “bad growth ratio”, the bigger the difference.
Delinquent balances have been growing at a faster rate than current balances, but this ratio has been falling — in other words, improving — over the last year, continuing the previous year’s trend. This improvement has taken place for delinquent balances at 1 cycle, 2 cycles and 3 cycles.
For example, in November 2013 the growth rate of 3-cycle balances was 1.63 times the growth rate of current balances. In November 2014, the ratio was 1.57.
The ratios for 1-cycle and 3-cycles reached their lowest point in more than two years. Year on... [Read More]
I guess I can pack up my desk and go home. My career was just ended by none other than basketball legend Charles Barkley, who dealt a death blow to the burgeoning analytics industry with this damning assessment:
It starts with Charles calling Houston Rockets General Manager Daryl Morey “one of those idiots who believes in analytics.” And it’s all downhill from there.
Joking aside, I think what’s telling about Charles’ diatribe is his emphasis on the difference between human experience (playing sports) and an abstraction of it (analytics). Charles isn’t alone here: There has always been a reaction to an over-reliance on analytics at the expense of human experience.
FICO experienced this way back in the 1960s, when founders Bill Fair and Earl Isaac would waltz into a meeting with bank risk managers and tell them that they weren’t making good enough decisions. The people charged with evaluating... [Read More]
Debt collection is at the intersection of two very popular news stories: the continuing shift of global wealth, and the rise of customer experience as the defining characteristic of successful companies.
Here’s where debt collection comes in. While the world’s richest are getting richer, many consumers have not. Many formerly affluent and middle-class people are struggling financially. Many of them are, for the first time, finding it a serious challenge to make timely payments on debt obligations.
Still, consumers expect every experience to be a great one – including collection efforts.
Banks and other credit issuers have gone to extraordinary lengths to create mobile apps that let customers deposit checks with just a photo, have enabled ApplePay, and more. Card providers alert consumers with a text message when they suspect a fraudulent transaction. Consumers love it. So why do so many debt collection efforts consist of calls from an automated... [Read More]