Credit scores and attrition: Correlation?

The quest to reduce the dreaded attrition rate remains a high priority for anyone in the monitoring space, and companies continue to explore new ways to predict and prevent cancellations. Companies don't just track attrition rates; they look for clues, like usage patterns, that could yield information about whether certain customers may be more prone to stay or go.

While usage patterns remain a valuable tool for evaluating customers and forming effective business strategies (conventional wisdom says upsell to active users, and reduce prices to the less engaged), it's not the only predictor companies use. There is also a significant correlation between credit scores, or Beacon scores, and attrition rates, according to Michael Barnes, a partner in the consulting and advisory firm Barnes Associates, who in a response on the CSAA's Accent forum, said his firm reviewed data on over 2 million accounts. Here's a bit of what he had to say:

“Generally speaking, the correlation changes over four ranges of scoring. Below 600, the statistical experience is very bad. That is, the accounts have a very high cancellation rate. Between 600 and 650 the results improve dramatically, with a general inflection point around 620+/-, which is why so many dealer programs (and, in some cases credit facilities) have restrictions around this area of scoring.”

Barnes added that scores above 700, in terms of attrition and retention, tend to behave the same as scores around 800. Scores in the ballpark of 650 tend to have poor cumulative performance, with the rates of cancellation almost twice as high over the first four years, Barnes notes. Unsurprisingly, rates of “infant mortality'—cancellations within the first year of existence—were exceedingly high among those with sub-650 Beacon scores, according to Barnes' data.

While the data sample is large enough to provide a thorough understanding of the relationship between credit scores and attrition, Barnes points out that some qualifications are needed, since a slew of factors can create exceptions. Some of these key variables include installation fees, services provided, pricing and payment method, and even geographic location.

The above graph, made for SSN in 2009 by the Edmonds Group, also charts the correlation between attrition rates and Beacon credit scores.