I’ve spent nine years of my career in the credit information industry at Experian and FICO. This article is not reflective of any scores by either of these companies.
A Kolmogorov-Smirnov (or KS) test is a form of minimum distance estimation used as a nonparametric test of equality of one-dimensional probability distributions used to compare a sample with a reference probability distribution. The KS statistic quantifies a distance between the empirical distribution function of the sample and the cumulative distribution function of the reference distribution. The reliability of credit scores is often measured by the score’s KS value. The greater the KS, the more predictive the score model.
Nice information for a credit manager to know, but does he really care? Most credit managers are more interested in identifying risky customers and taking evasive action before they become delinquent. A credit manager is interested in reducing outstanding debt, charge-offs and operation costs. Most credit managers couldn’t care less about how a credit score model is created or measured.
If I told my credit manager customers that my scores have a KS of 35 when my competitor’s model has a KS of 30, they would say, “That’s nice, but what does that mean to me?” If I told them that we can help them to identify customers that are going to become severely delinquent in their payments before it happens, they’re interested. If I tell them I can also help them reduce their operating costs by automating manual processes, freeing up staff to focus on particular customers all while increasing the frequency of credit reviews from annually to monthly or weekly, they’re very interested.
Credit managers aren’t averse to formulas by any means. You just have to speak in terms of the right formulas. DSO is a very important statistic for credit managers in manufacturing and distribution industries. Days Sales Outstanding indicates how much money stays out on credit for how long. The lower the DSO, the better.
The key takeaway is that it’s imperative that you speak in terms that are meaningful to your customer and focus on benefits they can relate to, not features that are important to you as a product manager or developer. You can use industry jargon all you want within the walls of your company’s office, but once you go out to speak to customers, be sure to speak their language.
OCPM's leadership team plus guest contributors share their insights on product management, product marketing, professional development & personal branding.