Machine Learning Fraud Detection Systems Could Save Banks $12bn Annually  

Adaptive behavioural analytics software reduces ‘genuine transactions declined’ by over 70% and incidence of undetected fraud by 25%

Oakhall, the London based analysis firm, estimates that global financial services firms could save at least $12 billion annually by employing adaptive, machine learning fraud management systems according to a study published in conjunction with Featurespace. For the full study see www.featurespace.co.uk/cost-of-card-fraud.

By employing adaptive behavioural analytics software to both identify actual fraudulent transactions, and reduce the number of ‘genuine transactions declined’ – as well as reducing the costs associated with managing blocked customers – the industry could reduce the $31 billion total annual cost of card fraud by over $12 billion annually.

Featurespace is a world leader in adaptive behavioural analytics software. Its services and products are employed in over 180 countries to a wide range of customers, including the leading US payments processor, TSYS, as well as Vocalink/Zapp, William Hill and Betfair.

Genuine transactions declined, also known as false positives, are legitimate transactions that have been incorrectly blocked by existing fraud prevention systems, which result in lost revenue and additional management costs to the card issuer.

By using adaptive behavioural analytics software, card issuers can reduce genuine transactions declined, improve operational efficiencies and lower the incidence of undetected fraud, according to the Oakhall study. Working with banks and cards issuers, Featurespace demonstrated a 25% reduction in the incidence of undetected fraud and, simultaneously, a 70% reduction in genuine transactions declined, as well as a subsequent reduction in call centre costs of 50%.

Oakhall applied these results to industry data to estimate the implied savings for the industry at $12.2bn, comprising $4.1bn reduced fraud and $8.1bn reduction of fraud management costs and lost revenue.

Jonathan Crossfield, Partner at Oakhall, said:

“Incumbent systems can block 10 legitimate transactions for every fraudulent one identified and, with undetected fraud, cost the global card industry $31 billion in losses, operational costs and revenue lost to competitors.

“A key benefit of adaptive fraud management systems is the much lower incidence of genuine transactions declined, with the potential savings of $12 billion annually.”

Martina King, Featurespace CEO, commented:

“Having genuine transactions declined is extremely frustrating for consumers and damages their relationship with the card issuer or bank. They also result in lost revenue and substantial costs to the bank. 

“Data-driven adaptive behavioural analytics – delivered via the ARIC engine – protects bank revenues and substantially cuts operational costs from false fraud alerts. It also helps the banks maintain positive relationships with their customers.

“The leading US payments processor, TSYS, chose to provide ARIC to its customers because of ARIC’s enhanced machine learning capabilities and decision-making around fraud and genuine transaction activity.”

Oakhall estimates that the total annual cost of card fraud to card issuers is $31 billion, of which $16.3 billion is associated with genuine incidents of card fraud, according to The Nilson Report, which compiles statistics on the global payment industry, and $14.7 billion are fraud-related costs associated with genuine transactions declined.

Featurespace’s advanced fraud management systems, that utilise deep machine learning and adaptive behavioural analytics, understand the ‘good’ behaviour of each customer more accurately and efficiently than fraud analysts, to reduce the losses and costs associated with managing fraud.

image002 image004

Author: Dylan Jones

Share This Post On
468 ad

Submit a Comment

Your email address will not be published. Required fields are marked *