In the light of South Africa’s recent greylisting, it might seem like one more nail in our economy’s coffin. And, with all the strategic actions to get back on track in the government’s hands, it can feel disempowering to have no control over what happens next. Whilst the majority of companies are not in a position to influence governmental action or policy, there are still steps that you can take to reclaim your power.
Where to start
Empowerment comes from making choices about your situation – and it helps to begin with understanding what that is and what you can do about it.
Greylisting is a result of deficiencies in a country’s anti-money laundering (AML), combatting the financing of terrorism (CFT) and proliferation financing (CPF) framework. The South African government has committed to The Financial Action Task Force’s (FATF) eight strategic steps and will be under increased monitoring whilst they are implemented. That aspect is out of your control.
Essentially, greylisting means that our AML / CFT systems are not robust enough, so there are too many opportunities for illicit money flows. Service providers in banking and finance, fintech, insurance, payment gateways, transaction switching gateways and voucher-based vending will all come under scrutiny. And this is where you can empower yourself by playing your part.
So, what are your options?
Understanding what’s possible
Using the right technological capabilities to analyse and visualise the complex data relationships in your transactional environment you can monitor interactions and identify patterns that indicate AML, CFT and CPF activities. Once confirmed, accounts can be blocked and the information reported to the authorities for further investigation and prosecution.
But what are the “right” technological capabilities?
There are various software suites using machine learning, artificial intelligence and big data analysis to detect dubious activity and mitigate financial risk. However, without the presence of a graph database to connect the dots and mine deeper, the ability for complex relationship analysis is limited. Graph data models allow you to track money flows, identify suspicious patterns that indicate illicit transactions (e.g. link prediction, similarity, clusters) and monitor millions of transactions in real-time.
The obvious choice
By prioritising the relationships (edges) between datapoints (nodes), graph reveals hidden patterns and trends that would be difficult to spot using traditional data analysis techniques. Assigning properties (e.g. time, date, amount) to edges increases the detail and value of this analysis, especially in recognising complex relationships between different entities. For example, allocating labels such as debit / credit or transfer to / send money allows you to run queries such as show all accounts with multiple debits or transfers in a designated time period. Support this with graph analysis of customer profiles, unstructured data and other relevant information and you can determine the presence and scale of any questionable dealings.
For additional context, graph models can also be primed with third party data sets such as Politically Exposed Person (PEP) lists, OECD sanctions data, Interpol Red Notice lists, public registries and leak databases.
Relational databases struggle to cope with the volume, complexity and dynamic nature of financial data structures. But graph’s ability to add or modify data without having to make significant changes to the database structure creates the flexibility to accommodate the massive volume of transactional data that is generated daily.
Then there’s query speed. Graphs are optimised for the type of queries used in graph analysis, such as pathfinding and pattern matching. Therefore, it stands to reason that graph is faster and more efficient than relational databases at performing complex queries on large data sets.
On top of all that, a graph database can be customised to your specific transactional environment. This results in a more tailored approach that can be integrated with existing systems and work flows.
So, it goes without saying that graph technology is the obvious choice for analysing the complex and hidden interactions produced by illicit financial flows.
However, a graph database cannot do all of this by itself. For a truly robust AML, CFT and CPF capability it requires an ecosystem of complementary software tools to optimise its potential.
Unleash the power of graph
With time of the essence, building a supportive ecosystem for your graph database inhouse doesn’t make sense. Which means there’s one more decision to make: Which third party vendor do you choose?
Locstat LightWeaver®, our graph intelligence platform, provides all the necessary technological components to place the power of graph in the hands of non-technical users and business analysts. Our easily configurable and rapidly deployable framework seamlessly integrates into your existing systems to deliver rapid ROI.
If all that doesn’t convince you, here are some more (though still not all) benefits:
- Feature engineering on data pipelines to curate and enhance datasets for advanced analysis.
- New use cases can be operationalised 90% faster to stay on top of changing circumstances.
- User friendly graph visual analytics that display the connections and flows between entities, parties, customers, accounts and transactions to facilitate rapid sensemaking within complex financial systems.
- A comprehensive rule database to facilitate rule analytics, auditing and building of dynamic rule models.
- Complex event processing (CEP) rules that are defined and built within an intuitive, criteria setting interface that requires no coding experience and can be tested for functionality and impact.
- Graph algorithms and models to combat fraud and money laundering.
- Operationalisation of GraphML models to help automate at scale.
- Extensive reporting capabilities (including aggregated and automated) which allow for customised reports (e.g. SARS, regulatory compliance) and can be disseminated via various channels.