Editor’s note: This is the second of three articles by Linda Straub Jones and Jeff Jones on the use of big data by credit unions.
Whether a credit union operates its own internal collections department or outsources those efforts, CUs understand that prioritizing collection accounts according to the likelihood of recovery is essential for cost-effective collections. While determining that prioritization is tricky, collections agents can increasingly rely on big data to fill gaps in their knowledge, to determine which accounts are worth the expense and effort of collection and which are not, and how to prioritize. Big data and collections analytics therefore can play a huge role in the outcome and profitability of collections activities.
Prioritizing accounts is important for a number of reasons, practical and otherwise. Collection organizations can substantially improve the effectiveness of their recovery and operate more efficiently if they know who is in a position to actually pay off their accounts and who isn’t, based on assets that might be invisible to the naked eye.
While collection via litigation took a dip after the recession, it is expected to see a rebirth as the market rebalances and industry regulations are clarified, and as consumers become increasingly able to repay their debts. However, the recession has left its mark, and older models of data collection that rely only on credit data to predict the likelihood of debt repayment may become considerably less reliable and less predictive.
Starting with gaining clarity about which defaulting consumers do and do not have assets allows collectors to focus their attention on the more fruitful accounts while also providing crucial information on whether to sue. Without big data, collections professionals run the risk of investing in a lawsuit against someone who ultimately can’t pay, wasting both time and money.
Big data can fill the gap left by the more traditional predictive models. Not only do data companies have the necessary information, but they also have the scoring and analytical capabilities to help in prioritizing collection accounts, using a scoring system that will predict the likelihood of debt repayment based on information about the borrower’s assets.
A scoring-based approach will help collections professionals to focus their efforts on accounts with a higher level of assets; align expenditures to maximize return on investment in each score band; optimize profitability with customized treatment plans in each score band and mitigate regulatory exposure on accounts with the lowest likelihood of repayment. As the market changes and traditional methods of recovery prediction can no longer accurately predict a return on loans, big data analytics can help collection organizations to substantially improve the effectiveness of their recovery and operate more efficiently.
