Credit & Lending

Customer Opportunities & Challenges

Across the many sub-segments of commercial and consumer credit and lending, there are some common themes in data analytics. One prevalent theme is that using a diversity of complementary and up-to-date sources can improve the fidelity and specificity of risk models used to make spot credit awards, and can also improve downstream monitoring of ongoing risk if the information that is being fed for ongoing analysis is kept current. There is substantial business value in better risk models. If a better job can be done projecting late payments or default, it helps the issuing companies minimize their financial exposure and be proactive in managing emerging customer situations, and it also improves the satisfaction of end-customers through improved pricing, more tailored offerings, and possibly obtaining credit or loans or mortgages when they might otherwise have been rejected based on limited and sometimes stale demographic (portfolio-level) data. 

Emerging online marketplaces, including in the subprime lending space, peer-peer and small business lending, and home mortgages, are adopting aggressive stances to supplant banks by providing easier awards which are made in real-time through online interfaces.  The award decisions are made with only limited customer (entered) data, by pulling and integrating additional information into the analysis behind the scenes. There is widespread interest to take advantage of more non-traditional sources of data, such as might be mined from utility payment histories, bank statements and transaction histories, social media sites like LinkedIn (job status) or Facebook (marital, medical), company websites, online purchase histories, mobile device behavior profiles, etc., provided this information can be used in accordance with fair lending practice. This would afford the more nimble actors an information advantage, and also allow them to identify more viable lending opportunities around the margins.  Overall, the online trends in credit and lending push attendant requirements for real-time analytics and automated decision making.

However, credit and lending customers may be challenged to realize those opportunities because:

• Accessing, pulling, aggregating, and analyzing data in real-time across diverse and silo’ed sources is difficult (hard to keep pace)
• Mixed staleness in the data used for modeling limits fidelity because data is evolving and sampled at different timescales (information turbulence)
• Managing the risks of unauthorized access, corruption, manipulation, and hacking is a costly and complex problem (compliance and security challenges)

Our Value Proposition for Credit & Lending

Collaborative Analytics (CA) offer an alternative method for leveraging data which is much more “arms-length” since the distributed data itself is never integrated, but is instead processed in place.  This presents customers in the credit and lending space some significant benefits:

• Makes it low cost and fast to exploit a variety of different sources together to make lending decisions, including external sources for which the raw data itself could never be accessed due to proprietary or privacy restrictions; new sources can be incorporated as soon as they come available without the cumbersome process of relearning an entirely new model from scratch
• Provides real-time automated prediction capability to sample risk profiles when needed, from the point of origination through subsequent servicing of loans and mortgages
• Provides an agile global data model which seamlessly integrates updates to local models as they are made, incrementally and on-the-fly as the system operates, helping to keep the models fresh and consistent with the latest structural changes in the data
• Reduces risks to privacy and hacking because the data are never centralized

Because the credit industry has a long history of modeling risk, most companies already have some kind of operational capability in place. A common scenario is that existing models are “expert system” or rule-based systems that are supported with manual intervention and oversight to “fuse” in other information and considerations, while internal development effort is being expended to build more automated capability based on machine learning and the latest methods in artificial intelligence.  A unique value proposition of CA is that it can augment existing models with additional sources of information, mined with the latest methods, without opening up the core model in use or even understanding its inner (proprietary) workings. This provides an extremely low touch way to try combining additional information to assess the benefit without disrupting the current approach in use.

Collaborative analytics provides a dynamic cross-silo risk assessment and monitoring capability that empowers credit and lending customers to keep pace with fast changing customer data.