Predictive Analytics paves the way for in-store credit in Africa

The rise of store credit cards in the United States has proven to the world that it is not only a manageable risk but, in fact, a hugely lucrative business model. This rise to success is owed, in part, to the effective marketing of super brand culture. Another reason is that in the post-modern landscape of superstores and global brands, a new currency has come to the fore – data.

Africa has its own fair share of home grown, as well and global, super brands and retail groups with enormous vaults of consumer data. Moreover, there is a bustling demand for micro-lending throughout most of the continent. Even mid-sized retailers with loyalty programs stand to capitalise.

Why is there hesitation amongst Africa based retailers to jump on the credit bandwagon? – Risk.

In the United States there is ample access to credible risk data for the majority of customers. Moreover they are far more credit worthy as a population, as a leading economy. This has allowed retailers to effectively manage the risk involved in offering credit options to their customers. A market previously limited to large, specialist financial houses.

In Africa, the reverse in generally true. Central risk indicators are sparsely available for the general population, making micro-lending excessively risky. Many play the numbers game and fall short, invariably seeking risk management advice too far down the line.

What do Africa based retailers have to manage credit risk? – Consumer data.

Traditional, central risk data providers use actuarial models that rely on an individual’s lending history; a luxury that is largely absent throughout much of Africa. With the latest Predictive Analytics, however, there is no need for that sparse lending data to build a usable risk model. Predictive Analytics makes use of patterns and trends in consumer data, such as customer purchase history and loyalty program logs, to reliably predict credit risk through spending behavior patterns.

While these models rarely match up to traditional lending based actuarial models, they are more than adequate for micro-landing applications such as in-store credit. Even where tradition credit score data is available, such scores are incorporated into the machine learning models and a more accurate combined behavioural-lending model is born, further reducing lending risk.

This is throws the door wide open for risk managed in-store credit offerings throughout the emerging African market.

Data is the new currency, and is redefining what is possible. Opening up markets that where recently out of reach and streamlining businesses beyond the abilities of human organisation. Machine learning based Risk Scorecard Solutions are just one of a rapidly emerging set of predictive tools that are propelling us into the future.

Find out more about the Risk Scorecard Solution here.