Spurred by a combination of factors, Lenders are actively looking to include more and different types of data in decisioning processes. Notably the fallout from lockdowns in recent years has exposed the limits of traditional data.
While exact models vary, traditional bureau credit scores consider the last 5 to 7 years of credit repayment history. This system advantages certain people, and in turn excludes many others. Traditional data leaves many lenders with no way to assess some people, cutting them off from potential good borrowers.
Alternative data is increasingly being included to allow lenders to ‘see’ thin file customers or to provide lift in the predictiveness of current processes.
Alternative Data could include other types of transaction data, such as bill payments. Or it could include character-based data – such as psychometric assessment or device data. Character-based data looks for character traits that have a proven correlation to willingness to repay a loan.
The explosion in data and the increased capabilities in AI & ML have opened many opportunities for credit scoring processes to benefit. Many factors influence the likelihood of someone to repay a loan. ML can better capture non-linear relationships which are common to credit risk.
There’s an increasing demand for lenders to show real change in their ESG responsibilities. For banks and financial institutions to succeed when it comes to ESG, there needs to be a much bigger focus on the social aspect. Catering to a diverse customer base means the need to review credit inclusion. The traditional approach to offering credit leaves millions of people outside of a system. The impact of excluding people from credit is significant. It drives people to use unsafe or predatory alternatives.
Open banking is an opportunity. It is a step in a journey of putting data back in the hands of borrowers. Liberation of data can drive greater insights and open opportunities for new borrowers.
Open banking has democratised the transaction data of millions of users around the world. There is a strong appetite to see how this data can be used for credit decisions.
When the pandemic hit, lenders had to change how they do business. 2 years later, the economic effects of Covid-19 continue to reverberate around the globe. The most financially vulnerable have only found their position becoming more precarious during the COVID-19 pandemic.
With a less stable credit environment, paired with a consumer market more willing to shop around, lenders will need to consider new ways to evaluate credit-worthiness.
While financial institutions want, and are being incentivised, to lend, they need to consider new ways of approaching risk assessment. Simultaneously, the shift towards digital, which has been steadily progressing for many years, has taken a giant leap forward. This push towards digital will see a reliance on leveraging new sets of data.
Analysts predict a trend towards cloud rather than on premise deployment, with an expected CAGR of 18.3% over the next decade. Cloud deployments are more economical, require less investment in terms of technical expertise.
Rather than a hurdle, clear regulations can actually be enabling. By informing customers of privacy policies and being transparent about how their data is being used, new credit assessment tools can enable consumers to leverage their data for their own benefit.
While privacy regulations vary in different regions, lenders can start by considering GDPR requirements, acknowledged as among the most comprehensive on the planet today.
In the past, purchasing and finance applications were distinct, separate processes. Access to data and advances in real-time analytics means that the credit check can now by carried out on the fly, as part the checkout process. Lenders can now tap new consumer segments at minimal marginal cost.
A lot of credit assessment mechanisms, such as traditional bureau scores, are based on third-party data. They consider data created from other transactions or activities and then build models to associate this data with likelihood to repay a loan.
There can be an element of volatility with third-party data as it is impacted by other forces that can change quickly and leave lenders exposed.
First-party data for credit is purpose-built. This means data created with the inherent purpose of indicating willingness to repay a loan. Character-based credit data, such as psychometric surveys are an example of first-party data. They are developed with the inherent purpose of creating data which is indicative of creditworthiness.
End users expecting more out of their customer experience and prepared to shop around. Customer loyalty is on the decline; consumers are up for grabs.
The changing nature of how consumers use credit, in particular the rise of BNPL providers, means that much debt is ‘hidden’ from traditional credit scores. While regulators are grappling with how to meet this change, lenders can reinforce their process by diversifying the data they consider.