Best fraud detection product: Clari5

Financial crime-management system uses collective intelligence and responds to changes in a client's behaviour over time

rivi-varghese-ceo-customerxps-1
Rivi Varghese, chief executive of CustomerXPs

When it comes to tackling fraud, banks have a built-in advantage, explains Rivi Varghese, chief executive of Bangalore-based financial technology firm CustomerXPs. For many of their customers, banks will have years of data, touching on every aspect of their lives.

"There is enormous collective wisdom they have about their clients, going back many years – but banks are not using this intelligence," he says.

Instead, they are mostly taking a "siloed" approach to fraud, with separate systems for their credit card business, debit cards, internet banking and so on. The failure of these systems to communicate with each other in real-time means banks are missing opportunities to identify and prevent fraud, Varghese says.

CustomerXPs's solution, the Clari5 fraud and financial crime-management system, is designed to break down these silos. The system not only responds to changes in a client's behaviour over time, but it can also use information from one part of the bank to identify suspicious activity in another function.

You have a system where if you invest money in your fraud-management system, you can make money out of it. As well as protecting the bottom line, it provides an opportunity to grow the top line
Rivi Varghese, CustomerXPs

For example, if a credit card payment is attempted in Greece shortly after an internet-banking transaction is carried out from a computer in the UK, Clari5 will block the payment.

This adds a layer of protection against hackers, who typically use what Varghese describes as a "segment" approach, where they attempt to hack into the accounts of tens of thousands of customers using one or two pieces of information. To avoid triggering a Clari5 alert, the hacker would need to have a much more rounded understanding of the target customer.

The challenge, from a processing point of view, is the ability to instantly synthesise and cross-pollinate intelligence from disparate systems in a split second. To do so, the Clari5 system needs to undertake complex analysis faster than the underlying systems.

To attempt this on ‘traditional' monolithic architecture would be prohibitively expensive. To dramatically reduce costs, Clari5 instead runs on distributed commodity hardware, aping the approach taken by search engine Google, says Varghese.

Anti-fraud network

The next step for CustomerXPs is to enable its member banks to share cross-channel and cross-business banking fraud information between themselves in real time.

The first phase of the Clari5 "federated anti-fraud network" is due to go live in September.

But the value of the Clari5 product does not stop at fraud protection. The holistic and real-time understanding it gives banks about their customers' activities provides opportunities to offer additional products or services – exactly when the client is most receptive.

"I know that a client is about to board an international flight, because they've swiped their card in the duty free shop at the airport and they don't appear to have bought travel insurance," says Varghese.

"You now have a system where if you invest money in your fraud-management system, you can make money out of it. As well as protecting the bottom line, it provides an opportunity to grow the top line," he adds.

Only users who have a paid subscription or are part of a corporate subscription are able to print or copy content.

To access these options, along with all other subscription benefits, please contact info@risk.net or view our subscription options here: http://subscriptions.risk.net/subscribe

You are currently unable to copy this content. Please contact info@risk.net to find out more.

Best use of machine learning/AI: CompatibL

CompatibL’s groundbreaking use of LLMs for automated trade entry earned the Best use of machine learning/AI award at the 2025 Risk Markets Technology Awards, redefining speed and reliability in what-if analytics

Most read articles loading...

You need to sign in to use this feature. If you don’t have a Risk.net account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an individual account here