Leveraging Graph Analytics to fight Money Laundering

It’s estimated that money laundering (ML) transactions account for anywhere between 2% to 5% of global GDP. This amounts to roughly $2 trillion per annum.

Today’s struggle against ML brings with it hefty fines to institutions found to have breached anti-money laundering (AML) regulations. Two giant cases were recently settled involving two banking conglomerates. Back in 2018, US Bancorp and the Commonwealth Bank of Australia were fined a total of over $1 billion between them for careless AML control and non-compliance offences respectively.

This ruling showed that no institution, no matter how big, will be afforded any leniency and that nobody is above the law when it comes to AML. Therefore, organisations need an effective, yet simple to implement AML system in place. But what are the challenges that impede these systems from coming into play?

Challenges to AML compliance

  • False positives – on average, 95% of transactions flagged as potential ML turn out to be false positives. That is a huge number of erroneous flags, resulting in more serious and potentially real ML attempts being missed.

  • Criminal technology – nowadays, criminals are getting more and more sophisticated in their attempts to launder dirty money. Technology has its downsides in that it is making things easier for lawbreakers to hide behind false names and complicated web identities.

  • Manual time-consuming investigations – turning a flagged item into a Suspicious Activity Report (SAR) takes between 20 minutes to a few hours to complete. This is because it takes time to dig into the multitude of records and networks which make each transaction possible. Given the huge manual time, this process is expensive and inefficient.

Luckily there are solutions which help in easing these burdens.

Graph Solutions for AML Compliance

Relational databases have until recently proved to be well equipped to support transaction monitoring and performing basic analytics. However, their one major limitation is not being able to identify relationships between datasets, which is an essential element when following money trails and assessing risks.

Fortunately, graph databases can perform all the above tasks – they can analyse relationships to reveal hidden networks and clusters which may be symptomatic of illegal activity. The way these databases work is rather unique in that they store data in nodes (customers and entities along with their attributes), thus creating connections.

For AML purposes, the money flow trails and parties forming part of ML transactions create a web of relationships. These relationships are created via the three phases of ML which are placements, layering, and integration.

Graph databases are a big help to the fight against ML. They bring fragmented pieces of AML data and combine them to create the story of a particular transaction or money movement including the means and systems used to process the money.

The future of AML compliance

A key aspect of combating money laundering activities is the clear representation of transactions in a manner which enables the understanding of the movement of the money in question through a number of accounts. These could potentially include several intermediaries. Graph analytics can also capture the means used to interact with the relevant accounts which can be utilised to further study the money transfer behaviour.

Graph analytics offers a unique solution to banking and financial services when it comes to AML efforts. There is a good argument to be made that this new technology will become the industry leader in the not too distant future and that it will be the expected and preferred technology by regulators.

An increasing number of financial service providers are investing in graph analytics for AML monitoring and compliance. They are recognising the fact that modern technology has turned the cumbersome AML compliance process into an advantage for their organisation. So, stay ahead of the pack and adopt this new technology now rather than tomorrow. Contact us to learn how ComplyRadar utilises a graph database, providing you with the power of true analytics to inspect and act on suspicious transactions in real-time.

What happens when your AML transaction monitoring rules are static

Although we live in a cutting-edge, modern tech world, a large number of payment service providers still use outdated, static systems to flag and identify suspicious transactions and payments. Remedying false positives can be a colossal waste of your team’s time. The situation can also lead to customers’ legitimate transactions being delayed while the team is handling and reviewing possible false positives. Given these facts, it is no surprise that a decade-old flagging system is no longer feasible for a number of reasons.

Too many flagged transactions

Most systems still use the static rules principle, which means that the system does not adapt to new rules but follows a stringent pre-defined list of criteria.​ This​ frustrating factor leads to transactions being flagged as suspicious because of their unique and modern ‘signature’. Obviously, if the signature is not found in the old processor’s list, it will mark it as a false positive, putting the expensive burden of review on the compliance team. Apart from being costly, a long review process increases operational costs, frustrates genuine customers and puts the company’s Anti Money Laundering (AML) compliance at risk.

Innovative criminals take advantage

Criminal exploits are made easier when old systems using extremely rigid rules are still being used. Static rule systems are slow to adapt, become quickly outdated, and have become increasingly predictable. These factors make it quite easy for criminals to hide their true intentions and mask transactions under the pretense of legitimacy. Likewise, with new rules and regulations put in place at an increasingly quicker rate, it is easy for an organization to fall out of compliance if it does not adhere to these latest laws. All this means increased costs through fines and updating of legacy systems.

Solutions to the problem

Fortunately, solutions do exist for these modern-day issues. For instance, AI-driven transaction monitoring enables organizations to stay ahead of the curve and always be compliant with new laws, while simultaneously improving customer experience. These systems have the capability to learn ‘on the job’ through continuous monitoring, inspections, and analytics – they actually get smarter. This means that now, criminal transactions can be identified more frequently and accurately as the system will have recognized the signatures used before and intercepted the transaction as suspicious. 

Internally, these new systems allow for a decrease in false positives and enable the compliance team to access real-time data involving legitimate behavior that is always changing. This further means that the systems are more cost-effective since manual reviews of false positives are eliminated. They also allow the compliance staff to focus more on the right investigations, thus creating an optimal AML workforce, unburdened by the constant manual reviews that hindered them in the past.

How ComplyRadar can help

ComplyRadar helps you address AML transaction monitoring requirements by automatically identifying suspicious behaviour in real-time or on a scheduled basis while minimising false positives. It monitors transactions related to individuals, accounts, and entities to detect suspicious activity quickly and effectively, through a fully audited process to inspect and act on flagged transactions. Contact us to learn how ComplyRadar enables you to fulfil your AML obligations whilst nurturing your genuine customers. It’s all about positive customer experience!