Data, Indispensable Lever Against Fraud

Beyond regulatory requirements, institutions must guard against the risk of financial fraud and any attempt at money laundering.  To face the ever-increasing inventiveness of fraudsters, data analysis enables to better understand and detect each malicious attempt. With better understanding of transactions, which the Logos ITS solution allows, fraud no longer has a chance.

Christos CHRISTOU – CCO at Lulu International Exchange

Beyond regulatory requirements, institutions must guard against the risk of financial fraud and any attempt at money laundering.  To face the ever-increasing inventiveness of fraudsters, data analysis enables to better understand and detect each malicious attempt. With better understanding of transactions, which the Logos ITS solution allows, fraud no longer has a chance.

All financial institutions are required to implement procedures to combat financial fraud and money laundering. Regulators around the world are increasing the requirements of financial service providers, who are forced to identify the ultimate beneficiaries of sums invested or transferred and trace the origins of the assets in question to ensure their probity. “We are therefore required to investigate what is behind each transaction going through our service and identify the actual destination of the assets we are processing,” explains Christos Christou, Chief Compliance Officer at LuLu International Exchange, active foreign exchange company based in Abu Dhabi. “Today, we must ensure that we are more equipped to combat fraud attempts in order to protect ourselves from an ever-growing regulatory risk.

“In order to improve the fight against fraud, we require a better view of the activity overall, through data analysis.” Christos Christou – Chief Compliance Officer at LuLu International Exchange.

A Better Understanding of What is Hidden Behind Each Transaction

This is not a simple task. Especially since those who attempt fraud can be very creative. They have numerous means to achieve their ends.  “We can distinguish between internal fraud – the manipulation of people within teams with authorisations to override controls – from external fraud, which is expressed by attempts at breaking into our systems,” states Christos Christou. “To fight better against any attempts to deceive the systems or hijack our procedures, we need to implement different approaches, and particularly, have a better view of the overall activity, connections and transactions that pass through our tools and services.”

Naweed Akbar Jalal – Compliance Supervisor & DeputyMLRO for the Barhain Financing Company

This is why data analysis is an essential lever for fighting better against fraud and money laundering attempts. “All the data at our disposal – account number, the amount exchanged, origin of the assets, their destination, currency, beneficiary, etc. must enable us to have a global view, better understanding of what is going on, and the intentions behind each transaction,” states Naweed Akbar Jalal, Compliance Supervisor & Deputy MLRO for the Bahrain Financing Company. “A structure such as ours deals with asset transfers. We do not manage client accounts. We therefore do not necessarily have a complete set of data with a history of transactions related to an account, for example. We must however meet very high standards and obligations in the area of fighting against money laundering, which requires us to find the means to better understand the intentions behind each transaction.”

 

“We broke the existing silos to be able to cross structured and unstructured data better.” Naweed Akbar Jalal – Compliance Supervisor & Deputy MLRO at the Bahrain Financing Company

Breaking the Silos

In this context, financial players must therefore establish advanced data analytics. LuLu International Exchange and the Bahrain Financing Company implemented the tool developed by Luxembourg company LOGOS ITS, which is today recognised as one of the best data analysis platforms to fight against financial fraud and money laundering. “Around this tool, we have implemented a central data management, to better analyse and understand it. We broke the existing silos to be able to cross structured and unstructured data better, whether it comes from within or outside the company. In this way, we have a broader view, allowing us to better identify contentious behaviours,” explains Naweed Akbar Jalal. “Since the tool has been implemented, we have detected many more suspicious cases that we were not aware of in the past. Such a tool greatly facilitates the work of the compliance team by using defined models that make it possible to trace suspicious cases.”

“We have a view on the logic at work. We can decrypt it more easily.” Christos Christou

Understanding Data from Different Perspectives

Thanks to the visualisation tools offered by such a platform, we can understand the transactions made better, and connect them to one another. “We have a view on the logic at work. We can decrypt it more easily,” explains Christos Christou. “In case of an alert, it is interesting to be able to understand the data from different perspectives, better understand the meaning and stakes behind each transaction. Today, without data, it appears evident that we could not do our job. It is up to us to ensure that it is of high quality by raising the awareness of the teams, and to always refine the models to control the risks and carry out ever more specific analyses.” With such tools, the compliance department can save a lot of time. “The investigation of each suspicious case is greatly facilitated.  While in the past, we would need a day to investigate a transaction, we now have a better view of the available data and can trace the information in half an hour. We therefore save a lot of time,” says Naweed Akbar Jalal.

“In the past, we would need a day to investigate a transaction. Now, we can trace the information in half an hour.” – Naweed Akbar Jalal

Considerable Time Saved

Better understanding of transactions through data analysis can therefore significantly reduce risks by better detecting suspicious cases and processing them more quickly. The time saved enables to investigate more complex cases better and reduce costs related to compliance procedures considerably.  The return on investment is directly noticeable. “According to current assessments, up to 70% of a data analyst’s time is devoted to false positive alerts or low value-added tasks. Only 30% of this time is devoted to real analysis work,” explains Olivier Merlan, Deputy Managing Director of Logos IT Services, publisher of the software. “With the right technology, you can simply reverse the ratio. The savings are huge. The challenge is therefore to improve detection by reducing false positives and allowing teams to focus on alerts that require further investigation.

The tool obviously adapts easily to regulatory changes. It allows for the implementation of a detection strategy adapted to each business and each situation. This strategy can be honed over time, according to the business outlook and apprehension of the risks to the company. The software is equipped with learning capabilities to gain precision over time.

Toward Predictive Models

The more financial institutions’ detection capabilities are strengthened, the more fraudsters change their practices, and the more their methods become complex. “We must therefore be able to follow them, and for that, cross an ever-increasing number of data,” explains Christos Christou. “New models must be developed, allowing in particular for preventive detection, based on an analysis of past cases, which make it possible to detect fraud risks in a given context.”

“We are developing models that rely on real-time global analysis and have learning capabilities that make it possible to identify fraudulent situations better and block them automatically.”

François Gourdon, CEO of Logos ITS

Today, we are studying the possibilities offered by data analytics to establish even more efficient models. “We are exploring the possibilities offered by technology to become more effective in the fight against fraud,” said François Gourdon, CEO of Logos ITS. “More specifically, we are conducting several research and development programs in collaboration with the University of Luxembourg, to provide better answers to the issues raised, namely by exploring the possibilities offered by machine learning and artificial intelligence. We are developing models that rely on real-time global analysis and have learning capabilities that make it possible to identify fraudulent situations better and block them automatically.