Machine learning and financial institutions

How is machine learning transforming financial services?

Standing still is the same as falling behind in today’s fast-changing financial services industry. At the heart of the transformation are new analytics techniques that help companies understand consumers more deeply than ever before: artificial intelligence and, in particular, machine learning. Seventy-nine percent of financial institutions globally believe artificial intelligence will transform not only the way they interact with their clients, but also how they collect, process and use their information.¹

What is machine learning and how can financial institutions use it?

In simple terms, machine learning is the ability of a machine to learn from data, without being explicitly programmed to do so. The algorithms derived from machine learning help find efficiencies in banks' internal processes, especially in areas such as portfolio segmentation, process improvements, and line of credit optimisation. For example, machine learning can help banks:

• Identify patterns of suspicious activity and proactively spot it
• Identify high-risk transactions before they are authorised, and prevent fraud attempts
• Identify sudden changes in consumer purchasing behaviour
• Minimise human error by automating highly repetitive activities
• Design more convenient offers and service channels for consumers
• Generate revenue with prediction models

However, identifying viable machine learning projects is a challenge for most organisations. To help banks take advantage of the opportunities that machine learning presents, Visa Consulting & Analytics (VCA) has developed new solutions to improve their portfolio profitability and performance.

AI in action: how data driven intelligence drives marketing

VCA provides its banking clients with effective cross-border solutions utilising machine learning techniques applied on rich global transaction data. The solution identifies:

1. Likelihood of a cardholder to travel in the next 30 to 90 days
2. Top destinations the cardholder is likely to visit
3. Cardholder segments most likely to be receptive to marketing messages to use their Visa card more

The solution uses machine learning algorithms to analyse more than 500 variables from VisaNet, Visa’s global processing network. These include total spend, spending frequency, average price and money spent across merchant segments both in country and abroad. With this data, Visa is able to create predictive models to understand the likelihood of a cardholder to travel in the next 30 to 90 days and the top three destinations they are likely to visit. The models let the bank customise relevant travel related offers to customers at the right time.

We have over 50 banks across the globe using this solution. To see the impact of the solution, we tracked six marketing campaigns using this solution. Banks saw an average 14.5% increase in cross-border travel spend per card, a 9.1% increase in cross-border travel transactions per card and a 7.2% increase in distinct merchant segments seeing cross-border travel purchases.²

Conditions for success

While machine learning is a powerful tool, banks must figure out whether it is appropriate for their current data configurations. For example, are their databases optimised for it? Does their information and technology infrastructure support the necessary tools? Is there a pipeline in place that can take machine learning model results into production? Perhaps most significantly, they must ask whether they have a team that understands both the underlying technologies and the nature of their business. Deriving value from machine learning requires not only processing information more effectively, but also extracting insights from it that can improve decision-making.

VCA consultants combine unique access to VisaNet, one of the world's largest payment networks, with advanced analysis techniques like machine learning and extensive experience in the payments industry. Together, these capabilities offer clients innovative solutions that catalyse transformative business growth.