With higher delinquencies and loan losses, along with tight budgets and more troublesome accounts, it is critical to know how to rank and assign accounts to be worked in collections.
Is your collection platform restricting the ability to leverage multiple data sources? The question should be top of mind for any financial institution.
This is a time when doing more with less is a must. Banks, debt buyers and contingency agencies need solid processes in place to appropriately evaluate and assess the risk of their portfolios - and to best determine those with a higher probability of payment.
In the last two years, the effectiveness of using information solely from the traditional credit bureaus has lessened because the meaning of this data has changed.
Traditionally, having a mortgage was viewed as one of the most credible signs of a borrower’s ability to pay. Today, that mortgage may be underwater or in foreclosure, restricting one’s ability to pay. A foreclosure indicator paradoxically actually may predict an ability to pay because mortgage payments are now halted and borrowers shift their priorities to credit cards to keep their credit lines open.
Scoring solutions must be able to adapt to a changing economy. With the recession, much can happen in one year or even a few months. Conventional credit bureau scores alone only tell part of the story. There are flexible and efficient new technologies, however, to address this concern.
Technology exists that enables sophisticated segmentation of the collection population and better performance assessment of individual accounts. Legacy risk management systems that require months to access new data sources or implement scorecard models can put financial institutions at a competitive disadvantage.
Today’s budgets can’t accommodate wholesale system changes. Inflexible legacy systems impact an organizations ability to quickly respond to market changes in risk and consumer behavior.
As noted by Dale Williams, president of Teletrack Inc., “Debt buyers and collection agencies now consider data from additional alternative data sources an important component to improve the efficacy of collection strategies and models. These data sources are incremental indicators not only of payment performance on financial services products typically not reported to the traditional credit bureaus [such as propensity to pay], but also of the consumers’ total current debt burden and payment obligations [such as ability to pay].”
Using credit data is no longer enough, as access to income verification, propensity sources, capacity indices and additional sources can establish a more predictive model. Non-traditional data offers a better strategic assessment of the accounts.
“Amid a rise in credit losses, collection models built solely upon established credit performance data may not be adequate to prioritize collections or predict collectability,” says Williams. “Non-traditional credit performance data can supplement existing models by providing insight into the frequency, type and payment performance of alternative credit that a consumer is accessing.”
According to Canh Tran, chief executive of Chicago-based Pattern Recognition, a predictive analytics and data mining company, innovative scoring solutions often leverage three components.
“One, newer scoring models make use of alternative data sets including non-traditional credit, social media and location-based information. Two, modelers use efficient data management tools to rapidly access, aggregate and assimilate the data for production. Eighty percent of the effort in getting a scoring model into production is getting the data right,” says Tran. “Three, non-linear data-mining technology applied to complex real world data sets can yield outstanding scoring results in portfolio pricing and recovery accuracy.
“In the ultra competitive collections world, companies that use advanced scoring methodologies in conjunction with credit bureau information will have a distinct advantage over those that only rely on widely available credit scores. It is like playing poker knowing the other guys hand,” adds Tran.
By leveraging the value of non-traditional data sources, a more holistic view of a consumer’s ability to pay can be developed and used. This will drive more effective collection and recovery strategies that can optimize the use of internal collection staff, third-party agencies and debt sales.
Integrating new technology that can deliver rapid access to these alternative data sources, as well as shorten the implementation timeframes for new models and strategies, only further enhances the value of past technology investments. The result will be a better-balanced workforce, as well as collection queues for the most effective decisioning, thereby reducing expenses and workload and accelerating recoveries.
Paul Greenwood is president and co-founder of GDS Link LLC, a provider of customer-centric risk management and process automation solutions.