Machine learning supporting customer experience

When incorporating innovative technologies to future-proof your business, it is important to recognise that when handling data, the incorporation of artificial intelligence (AI) and its subset machine learning (ML) is a hallmark of the cloud model.

As well as providing the capability of scaling up and down quickly to test and learn, cloud can train algorithms and improve predictive models to avoid false transaction declines, prevent fraud, and deliver on prototypes and proofs of concept quicker.

STAYING AGILE AND SECURE

While banks are operating with faster speed and greater agility in the payments space, shifting payment practices and buying behaviours have also drastically changed. With the rise of ecommerce, the number of card-not-present transactions has drastically increased.

And while the shift to digital banking was already taking place before Covid-19, the health crisis catapulted the adoption of contactless payments forward. With these accelerated innovations, the opportunity for fraudsters has also increased, creating expensive, long-term risks for financial institutions.

PSD2 in Europe mandates transaction risk analysis (TRA) to combat fraud when the customer is not physically present, but banks must be proactive, remain a step ahead by analysing both historical and real-time data and use ML to differentiate between legitimate and fraudulent payments.

Machine learning algorithms help to:

  • detect and prevent potential fraudulent transactions from being approved
  • reduce the number of false positives
  • decrease friction
  • help reduce complexity
  • make sense of emerging fraud patterns and correlations
  • answer questions that would have taken humans too long to answer
  • counter increasingly different and evolving forms of fraud

Banks must build and train more targeted, precise machine learning models that reduce financial risks to help protect themselves and their customers. When running critical and sensitive workloads, banks activate a broad suite of machine learning tools and optimised payment fraud detection algorithms and with identity authentication and biometrics, financial players and their customers are better protected from fraud.

Banks, processors, and large payment networks have ramped up fraud prevention utilising structured and unstructured data to support behavioural biometrics.

However, banks must strike a balance between protecting the customer and providing a good customer experience — this will continue to be an uphill battle. Friction can cost the bank money and impact customer loyalty.

Institutions globally are not going to move backwards to legacy mainframe- based systems; the digitisation of financial services will accelerate catalysed
by the pandemic. Issuers will increasingly move toward newer cloud native technologies over time that leverage the talent base prevalent in today’s market, increasing resilience and compliance in light of the regulatory requirements in the marketplace now and in the future.