Sexy job, sense of humor, slogan
By Peter Horner
They already have the sexiest job of the 21st century according to a Harvard Business Review article by Tom Davenport and D.J. Patil. Now, it turns out, data scientists also have a great sense of humor. Who knew?
The Institute for Operations Research and the Management Sciences (INFORMS) will hold exams for its Certified Analytics Professional program according to the following schedule:
March 6, 2014
Drexel University James E. Marks Intercultural Center
March 29, 2014
INFORMS Conference on Business Analytics and O.R.
Westin Boston Waterfront
March 30, 2014
Gartner BI & Analytics Summit
Venetia Resort Hotel & Casino
Las Vegas, NV
April 15, 2014
Queens University School of Business
Toronto, Ontario, Canada
May 22, 2014
University of Cincinnati
Lindner College of Business
June 21, 2014
INFORMS Conference on The Business of Big Data
San Jose Marriott
San Jose, Calif.
To apply, click on https://www.informs.org/Certification-Continuing-Ed/Analytics-Certification/Apply-for-Certification
For more information, click on https://www.informs.org/Certification-Continuing-Ed/Analytics-Certification
March 30 - April 1, 2014
2014 INFORMS Conference on Business Analytics & Operations Research
June 22-24, 2014
2014 INFORMS Conference on the Business of Big Data
San Jose, CA
Industry NewsAIMMS hosts EMEA Partner Days 2014
AMMS, an algebraic modeling language and software system, recently hosted its annual partner conference for Europe, Middle East and Africa (EMEA) partners in Amsterdam. CEO Gijs Dullaert kicked off the event by sharing an ambitious 10-year plan to bring AIMMS to the next level. AIMMS’ PRO platform and supply chain focus will be the basis for its growth, furthering AIMMS’ mission to bring the benefits of optimization to society.Read More
Special ArticlesINFORMS Big Data Conference
How do you get from data discovery to return on investment and real business value? The INFORMS Big Data Conference, set for June 22-24 in San Jose, Calif., aims to help you discover just that. This newly launched topical conference will put the focus squarely on the business of big data. A major component of the conference will be case studies of big data projects that illustrate the complete journey from business problem to analytics solution.Read More
CAP NewsContinuing education for analytics professionals
INFORMS’ popular continuing education courses, “Essential Practice Skills for Analytics Professionals” and “Data Exploration & Visualization,” will both be held prior to the 2014 INFORMS Conference on Business Analytics & Operations Research in Boston. Register early to save on registration.Read More
How to leverage analytics to build collections into a competitive advantage.
By Vishal Ahuja (left) and P.S.S Moorthy (right)
Credit cards and consumer and business loans have been experiencing vacillating delinquencies and losses over the last few years, resulting in an overall increasing trend in delinquency and write-off rates, as shown in Figure 1. In these turbulent times, staying focused on executable practices with a strong track record of results provides a guide for how companies are turning collections into a competitive advantage. The most agile, novel and flexible will succeed in reducing delinquencies. They will mitigate losses during this time by understanding customer behavior, customer communication preferences, and the effectiveness of various channels in increasing response rate and maximizing recovery.
Role of collections
Advanced collections strategies allow organizations to go deeper into a highly competitive marketplace in search of new business. An organization with a strong collections capability can gain a strategic advantage over the competition by being able to accept riskier customers without corresponding increase in delinquencies. This is done by understanding that not all delinquent accounts are the same. Some are cured and roll back to current, while others charge-off in due course. The explanation for these differences can be attributed to various reasons, e.g. adverse selection while booking new accounts; customers facing financial difficulties or mismanaging their finances; a bad economy; and, last but not least, customers just forgetting to mail the payments. Organizations need to collect on higher risk accounts as soon as possible while giving lower risk accounts time to selfcorrect, resulting in more collected dollars with reduced collection costs and chargeoff losses.
Figure 1: Increasing delinquency rates across all loan groups (Source: Federal Reserve Board).
Usually a customer has multiple credit relations, which means the share of a customer’s wallet that you could get depends on what you do vis-à-vis your competitors. If you do not collect soon enough, you risk not receiving a payment, as your competitors would have collected all that the customer had to offer. On the other hand, if you start collecting from the day an account cycles in to delinquency, you may end up irritating some good customers and lose their business. In addition, this would be a misallocation of limited collections resources.
Organizations need to:
- Contact the right accounts, quickly and frequently. Identify accounts with higher probability of losses and work on them proactively rather than waiting until they turn non-earning.
- Improve productivity. Avoid collection calls unless necessary; they cost money.
- Use workflows that suit the company’s needs. Have the right strategy (phone call/e-mail/voice mail) for each account, using the power of information to your company’s advantage.
- Close loop with risk teams. Inputs from collections teams will help strengthen the underwriting criteria.
These four objectives can be achieved by the right application of analytics. Analytical models can predict the likelihood of payment and can be used to reduce collection costs by prioritizing debtors and delaying more expensive collection efforts for those most likely to make payments. This section describes the analytical approach in designing targeted collections strategies that will improve the effectiveness and efficiency of collections.
Broadly speaking, the collections life cycle includes three stages depending upon the delinquency associated with the accounts, and the objective of the collections team in each stage is different as illustrated in Figure 2.
In the early stage of delinquency, the percentage of self-cures/self-pays (accounts that pay on their own without any collections effort) will be higher due to specific behavior patterns such as paying on a particular date every month before or after the due date. Analytics can help identify the characteristics of self-cures, thereby helping the collections team to focus their collection efforts on those customers who need it the most. Different scoring techniques can then be used to segment the remaining delinquent accounts into high, medium and low risk segments.
The desired objective of each segment is different and so should be the strategy deployed. The collections goal for a high-risk delinquent customer is generally to recover as much of the money owed as possible within an acceptable timeframe. Further segmentation of this group can help the collectors segregate those who may be retained as good customers following a change in loan/lease terms, hardship plan or other renegotiation. Similarly, in the late stage of delinquency, it is important to maximize the collections by making the customer pay as much as he can. It is also important to evaluate and identify specific actions that are more likely to yield good results during this stage. The decisions to be made in either case are very similar:
- When do I contact a customer?
- Which customers do I contact first?
- How do I contact a customer?
- What do I say to a customer?
Determining who, how, when and what to say is what predictive collections analytics does. It predicts the probability that an actively worked debtor will make significant payments. Some debtors will make significant payments with minimal, inexpensive efforts; others require more intensive efforts. This in turn will dictate how many calls need to be made, how many agents are required, etc., as shown in Figure 3.
Figure 3: Collections analytics provide answers to some key questions.
Mailers: Analytics helps in evaluating the trade-offs between contact strategies targeted at specific segments. Standardized techniques are used to identify the most successful and optimized treatment for each segment of customers based on their historical response behavior for different kinds of treatments and the stage of delinquency (early/late). For example, certain segments of customers might respond more readily to technologies such as an online chat or text messages or automated voice mail compared to a phone call. In some cases, special mailers can be sent with different verbiage for offering payment plans and settlements offers.
Best time to call: By analyzing historical outbound call and response data, the best time to reach the right party can be determined, which enables collections to initiate phone calls to delinquent customers at the times and places they are most likely to be reached. This improves contact efficiency by increasing right-party contacts and focusing calling efforts on the high-risk accounts.
Prime time is the period of highest inbound activity and outbound productivity. Analytics helps in developing and deploying collection center staffing models that result in reduction of staff size, as well as expanded collector scheduling during prime time periods.
Optimal penetration: Analytics plays a vital role in reconfiguring dialer strategies through a champion-challenger framework to arrive at optimal penetration rates. Contact rates are positively correlated to the number of collections attempts, but the law of diminishing marginal returns eventually sets in and any incremental attempts will result in few contacts. Analytically driven penetration targets are necessary for optimal staffing and collections efficiency.
Optimal abandon and idle rates: Through predictive dialing, non-productive calls are filtered and not sent to the collectors, thus enhancing performance and results. A higher line-to-agent ratio will result in a higher abandon rate but a lower agent idle rate, and vice versa. Analytics can help achieve optimal dialer pacing, which provides maximum contact rates and minimum agent idle time.
File rotation: Calling strategy should ensure that delinquent customers are not contacted at the same hour every day. In a predictive dialing scenario, spreading collections calls adequately throughout the week maximizes the chances of right-party contact. It also prevents the likelihood of training the delinquent customers on when to expect a collections call.
Execute collections operations
Incentive: Invariably, collector behavior will align with the organization’s incentive structure. If the organization rewards efficiency, then agents will stress minimum payments. Conversely, if incentives are linked to dollars collected, agents will ask for the full amount due. Analytics can assist in building incentive mechanisms that will mitigate the conflict between an agent’s best interest and the organization’s.
Scheduling staffing: Staffing a collections center is no simple matter, and adding in the complexities of the collections business process makes the task even more difficult. Analytics helps organizations schedule the right agents, with the right skills at the right time to deliver high-yield campaign results, while maximizing the efficiency and effectiveness of staffing.
Analytics can add considerable value to an organization’s collections operation in terms reducing delinquency, mitigating losses, enhancing collection efficiency and reducing overall cost of collection through a broad range of solutions.
Vishal Ahuja is vice president of operations for Genpact’s collections analytics practice. With an extensive background in the retail finance industry with specific experience in credit, collections and fraud operations, he has build marketing, risk management, collections and fraud prevention strategies. He holds a master’s degree in quantitative economics. P.S.S. Moorthy is a manager with Genpact’s analytics business. He has extensive experience in the field of collections analytics, originations analytics and transitions & quality. He currently works on commercial collections analytics. For more information on Genpact, visit www.genpact.com/home.aspx.