Global financial institution targeted “un-lendable applicants” by leveraging previously-siloed internal and external datasets to increase nuance in risk scoring.
The self-employed – an essential pillar of the economy – traditionally struggle to access key financial products because they don't conform to traditional credit scoring criteria.
Referred to as ‘thin-file’ customer segments, these solopreneurs slip through the cracks as current risk models can’t accurately score them.
In order to better serve this overlooked and underbanked customer segment, a leading European bank wanted its evergreen digital offering to combine the very best of its traditional banking expertise with the latest technologies.
By partnering with SparkBeyond, the bank could identify accurate micro-segments (gig economy workers, students, entrepreneurs), quantify individual risk, and tailor products with deep personalization for customers on the hunt for improved offers.
AI-powered analytics has become an essential part of the digital toolkit for innovative banks looking to measure credit risk and meet lending criteria for previously un-lendable applicants.
Technology, like that offered by SparkBeyond, applies AI-automated credit risk scoring by taking vast amounts of internal bank data, however scarce or fragmented, and augmenting it with external datasets to find unexpected correlations. This external data ranges from mobility, geospatial, demographics, healthcare, telematics and IoT, through to census, politics, trade and economic indicators, and climate change.
By leveraging ‘world knowledge’ data and testing millions of hypotheses per minute, AI reveals previously hidden root causes behind customer default and identifies the macro-economic pressures likely to affect future behaviour – giving lenders a holistic picture. New insights are then used to generate fast, accurate credit risk assessments of non-traditional loan applicants.
Revolutionary credit scoring like this eliminates human decision-making bias and makes loans available to millions of people with patchy or unverifiable credit history, ranging from sole traders and smallholders to gig economy workers and students.
Using its AI-powered platform, SparkBeyond automatically combined six of the bank’s previously siloed internal and external datasets, and revealed 50 million patterns driving risk.
The bank then used these insights to successfully launch its first venture: a unique card+app service to help solo entrepreneurs manage their private and corporate finances in a single place. This product is now serving a growing microsegment – solopreneurs who build and run their business alone, and have traditionally been underserved by banks.
Apply key dataset transformations through no/low-code workflows to clean, prep, and scope your datasets as needed for analysis
Apply key dataset transformations through no/low-code workflows to clean, prep, and scope your datasets as needed for analysis
Apply key dataset transformations through no/low-code workflows to clean, prep, and scope your datasets as needed for analysis
Apply key dataset transformations through no/low-code workflows to clean, prep, and scope your datasets as needed for analysis
Apply key dataset transformations through no/low-code workflows to clean, prep, and scope your datasets as needed for analysis
Apply key dataset transformations through no/low-code workflows to clean, prep, and scope your datasets as needed for analysis