Digital Footprints is the information users of a website leave when they access any website or app online or register online.
Berg et. al (2020) examine how digital footprint are used for credit scoring by fintech intermediaries or lenders. This approach is different than traditional techniques. There is no scavenging of data or information sought from borrower, related to financial information, or social network data. Examples of digital footprint:
- Device used - Desktop / Mobile
- OS used - Windows, Android, iOS
- Source of origin of customer visit to the commerce site - Search / Ads
- Whether email id provider is a paid service or free service
- Time of the day when the purchase is done
Berg et al (2020) suggest that digital footprint proxy for income, character and reputation and can be used for default prediction. Some of the individuals may not have standard income documents or credit scores. "For example, the difference in default rates between customers using iOS (Apple) and Android (e.g., Samsung) is equivalent to the difference in default rates between a median credit score and the 80th percentile of the credit bureau score."
Customers coming from a price comparison site are less likely to default as customers being directed from a search engine, distinguishing consumers personality traits for impulse shopping.
Customers having their names in the e-mail address are less likely to default. The differences in default rates between a median credit bureau score and the 70th percentile of the credit bureau score.
The discriminatory power of identifying good cases using digital footprint is as good or better than that of credit bureau scores. Additionally digital footprint based methods have higher predictive powers and hence can also be used for traditional longterm or medium term lending products.
Berg, T., Burg, V., Gombović, A. and Puri, M., 2020. On the rise of fintechs: Credit scoring using digital footprints. The Review of Financial Studies, 33(7), pp.2845-2897.