Has the internet changed the nature of the way shopping occurs in stores these days? more
Facebook Places, Foursquare and just about every Social Networking seems to be focusing on location these days. But the problem is they all rely on some sort of check-in, whether it be manual check-in by the user or using the latest GPS/Geo-fencing based check-in.
The recent credit crisis highlights some of the inherent weaknesses in the current model of providing consumer credit. That model relies on tried and true inputs and limited real-time data. When the factors turn out to be wrong or incomplete, the models fail.
The introduction of new services can provide an opportunity to incorporate new data inputs into the old models, as well as, perhaps for the first time, real time information regarding buyers shopping patterns and interests. That new data could lead to a more dynamic credit scoring system that is edified not only by past behavior but also by current, real time activity and shopping pattern changes.
Using a model based approach, credit providers dole out credit to consumers based on multi-factor models that analyze past behavior patterns to determine not only suitability for, but also size of, credit lines. While the factors are constantly updated, they generally focus on past credit behavior of a consumer (how regularly has she paid her other forms of credit, how many credit lines does she have, current balances, last reported salary, etc.) rather than current or even future behavior. Simply, current data collection is difficult, all of the obvious (and many not so obvious) credit factors seem to be known and we have no way of seeing into the future. As an aside, only recently have we started to understand the correlations between heretofore seemingly unrelated credit factors (mortgage lender and zip code default correlations for example) – much less incorporate daily behavior into credit models.
But what if we knew how often a consumer visited certain types of retailers, how often he purchased per visit and observed how these relationships and visits changed over time? Would we be able to make new decisions? As an example, a consumer who lost their job may, relative to their past history, reduce their visits and/or purchases from high end retailers but may increase their visits to coffee shops and bookstores. Could we assume therefore that person, all things equal may be less credit worthy? Could we have made that assumption if all we knew was that they showed up less frequently at the high end retailer? Perhaps, but with this type of data we will have the ability to start thinking about the analysis. With the ability to see current behavior, we may be able to see a credit problem before it occurs. This type of real time data may help credit providers decided to increase or decrease the amount of available credit as they see changes in lifestyle before they observe changes in credit activity; especially if used in conjunction with more traditional credit factors.
As one correlates shopping patterns to the data otherwise used in credit scoring (outstanding lines, open to buy, etc) a fuller picture of the consumer can develop. Credit providers can understand not only payment patterns but lifestyle patterns to better understand current and future behavior. They can use this data to make insightful decisions both positive and negative in ways that have not been available.
What’s needed is a card that can collect and receive of this type of data instantly. Then we will know when someone is in our store, if they bought, what they bought, and over time we see how their activity changes. So from a credit perspective, that data may allow one to see not only the present but maybe the future too!!!
