Debt collector predicts payment outcomes using data analytics

Transcription

1 Debt collector predicts payment outcomes using data analytics EOS Group uses advanced analytics and automation to predict debt collection success, reducing time spent on administrative tasks by 70 percent Case Study Company profile Company EOS KSI Česká Republika, s.r.o.l Industry Financial services Country Czech Republic Employees 3,900 medical faculty and staff Business need EOS KSI needed a way to gather insights from the massive amount of customer data held by one of its clients about past-due accounts to help predict the collectability of certain debts. Solution The company implemented predictive analytics software to help create collectability scores for the client s debtors, predict whether litigation might be successful and optimize collection for some debtors. Benefits Cost savings through collection process optimization Time savings through automation of administrative tasks Ease of installation and use

2 Debt collection it s a dreaded phrase if you re a business owner and one that strikes fear if you are past due on a bill. But for EOS KSI Česká Republika, s.r.o., debt collection is an area of expertise. The company specializes in receivables management, offering debt collection and other services through 50 subsidiaries in 25 countries, and working with clients of all sizes, including internationally traded joint-stock firms, insurance companies, credit card issuers, publishers and freight companies. Collecting debts walks a tricky line. On one hand, businesses cannot survive if they allow their customers to get away with not paying their bills. On the other hand, businesses that attempt to collect debts too aggressively from delinquent customers can usually count on losing those customers forever and too many lost customers can hurt a business s long-term viability. EOS innovative use of analytics has saved time and money and increased the collection of bad debts. - Ivan Fibir, Call Center Manager, EOS KSI 2

3 The good news is that many debts can be collected relatively simply. The challenge, says Ivan Fibír, call center manager for EOS KSI, is to gain enough insight into debtors circumstances and characteristics. With the right insights, Fibír explains, EOS hopes to not only collect unpaid bills but also help retain more customers after the debts have been settled through good customer service. No one wants to lose customers if they can help it, because even people who have a late payment now could become a great customer later, Fibír says. If we do our job well, a customer with an unpaid bill could go from feeling anxiety at the sound of your business s name, to not only settling the debt but also singing your business s praises to family and friends. But insight is key. The massive amount of data held by companies about their past-due accounts can tell many stories: which debts are most likely to become delinquent; which delinquent debts are most likely to be collected (and which will simply be a waste of resources to pursue); when is the most opportune moment and method for seeking payment; and what is the best means of contacting a particular customer. But without effective data mining and analysis tools, unlocking those stories is impossible. EOS KSI had been using an internal solution to obtain some of this information about its client s customers, but the company wanted additional tools that could help it score debts and predict the likelihood of success for each step of the collection process. The new Previously, we did all of our analytical work based on gut instinct and individual experience. Statistica now allows EOS to automate analytics and transform our processes and these changes are now mission-critical within our business. - Ivan Fibir, Call Center Manager, EOS KSI 3

4 solution needed to support the decisionmaking process around whether to go to court over a particular debt, what phase of the process was best for collecting a debt, and whether a standard collection step should be skipped. The solution also needed to integrate easily with EOS KSI s existing in-house solution. EOS KSI deployed a new solution based on Statistica Data Miner, Statistica Advanced and Statistica Reporting Tables. Statistica Data Miner helps build the models that estimate the probability of whether a debtor will pay or not and specifies the optimal collection strategy for each debtor. Statistica Advanced offers data cleaning and data preparation, and also helps EOS gain a deeper understanding of its client s Debt collection it s a dreaded phrase if you re a business owner and one that strikes fear if you are past due on a bill. But for EOS KSI Česká Republika, s.r.o., debt collection is an area of expertise. The company specializes in receivables management, offering debt collection and other services through 50 subsidiaries in 25 countries, and working with clients of all sizes, including internationally traded joint-stock firms, insurance companies, credit card issuers, publishers and freight companies. Products & Services Software Statistica 3 debtors and their behavior. Statistica Reporting Tables automate daily reports from the call center for management. With the new solution, EOS KSI has access to advanced, userfriendly tools for creating data models that reflect the many different companies, regions and circumstances of its geographically diverse operations. Using information such as type of client, nature of debt, region and available contact information, these data models help EOS KSI predict whether certain debt collection processes will be successful or how they should be modified to improve the chances of success. Previously, we did all of our analytical work based on gut instinct and individual experience, says Fibír. Statistica now allows EOS to automate analytics and transform our processes and these changes are now missioncritical within our business. 4

5 We were able to implement the Statistica solution quickly and easily, using our own employees and our users find that the Statistica environment is intuitive and simple, even for nonanalysts. - Ivan Fibir, Call Center Manager, EOS KSI EOS can now use payment probability and debtor characteristics and behavior to drive the collection process. In this way, some debtors may receive a phone call after a bill is not paid and some may receive dunning letters, which simply state that the customer is overdue. Dunning letters typically progress from polite reminders to more strident demands for payment. Cost savings Statistica has helped EOS KSI streamline its debt collection process to save money. Instead of an intensive collection process focusing on all unpaid debts, EOS can zero in on debts most likely to be paid, greatly reducing call center and administrative costs. Time savings Using the modeling in Statistica allows EOS to shave some of the actions from the collection process. In addition, the company has been able to save 52 hours in administrative time each month by automating certain tasks that s equal to one-third of one employee s total workload. For example, specialists at EOS can now easily connect and access data using the Statistica Query functionality without needing to enlist the help of the IT department. In addition, Statistica has automated daily tasks such as call center report generation for management review, saving team members one hour per day. With the Statistica Workspace Environment and Reporting Tables functionality, report generation now takes only one minute. Model generation in Statistica is also completely automatic workflows in Statistica run the models every night in batch mode, computing probabilities for new debts and delivering specific values of prediction for each. These predictions can then be used the next day to assign steps and a timeline to the collection process for each debtor. Easy to install and use Given the power and the value of the Statistica solution, Fibír is impressed at how straightforward the deployment was and how userfriendly the tools are. Statistica provided staff trainings on how to 5

6 use Statistica as well as on the theory behind model building and specifically how it could benefit EOS. We were able to implement the Statistica solution quickly and easily, using our own employees, says Fibír. Our users find that the Statistica environment is intuitive and simple, even for nonanalysts. Tomáš Jurczyk, Statistica solution consultant, was a tremendous help during implementation. After analyzing EOS KSI s previous system for tracking the client s past-due accounts, Jurczyk helped EOS KSI select what data to focus on with the new Statistica solution and advised the team on client segmentation. Jurczyk also helped with models development, implementation and monitoring, as well as customization of the software for creating daily reports of call center productivity. EOS innovative use of analytics has saved time and money and increased the collection of bad debts, says Fibír. Now that analytics is being used within a specific segment of the debt portfolio, EOS intends on expanding the automated analytics within other segments of the debt portfolio and other segments of the business. About Statistica Statistica s advanced analytics, big data and IoT offerings provide you endless possibilities to innovate your enterprise. Whether it s uncovering the genetic basis of a disease, reducing hospital readmissions, mitigating financial risk, or ensuring procedural validation, Statistica enables organizations to transform in new and exciting ways. By embedding analytics everywhere and empowering a wider community of citizen data scientists, you ll accelerate innovation, improve customer experiences, and streamline your enterprise for the future. View more case studies at statistica.io/resources Statistica and the Quest Software logos and products as identified in this document are registered trademarks of Quest Software, Inc. in the U.S.A. and/or other countries. Other trademarks and trade names may be used in this document to refer to either the entities claiming the marks and names or their products. Quest Software disclaims any proprietary interest in the marks and names of others. Availability and terms of Quest Software, Solutions and Services vary by region. This case study is for informational purposes only. Quest Software makes no warranties express or implied - in this case study. March 2017, Quest Software Inc. All Rights Reserved. CaseStudy-StatisticaEOSGroup-US-AA-00000