Identifying emergent financial risk through complex connections


Emergent risk relies on more than one or two data points

We all know that data is a disparate and ever-growing resource that provides invaluable and often indispensable insight to underpin the success of global businesses. Financial Services is one sector in particular where data is relied upon heavily to help manage business critical functions such as financial risk. Many screening tools are currently deployed within the sector to look for patterns or previously identified behaviour within data that help to identify red flags.

Emergent risk
The combination of chemical elements often creates new emergent properties. Similarly, the combined analysis of multiple data points can help identify emergent risk that would not be visible in standalone data points.

The problem is that the analysis of single or low numbers of common data points can’t always be relied on to uncover suspicious activity. The combined analysis of numerous, complex and varied data sources will however often present a very different emergent financial risk story. Much the same as how emergent properties develop in chemistry, biology or the natural world. Individual elements are not always representative of the combined output.

Machines rely on human traits to help identify emergent risk

Machines and AI continue to provide us with the tools to more rapidly and accurately interrogate volume data and search for ever more complex patterns and trends that may help to inform high risk decisions. Indeed, they are even able to make the decisions for us. It’s not quite that simple though.

Every risk-based case investigation conducted in the financial services sector can differ to varying degrees. The best risk analysts rely on their higher order intuition, knowledge and individual traits to identify or investigate red flags across differing cases. It is that kind of knowledge that machines require if they are to be successfully trained, managed and continuously optimised. Especially if they are to be trusted and accurate.

These factors are an integral part of Entity Investigator which can analyse high volume complex and disparate data points to provide risk assessments and advice for investigations teams that would otherwise take hours to undertake. More importantly it thinks using expert analyst traits and can therefore identify those emergent risks that often only become present with a combination of multiple data points. Without technology those points would be difficult to analyse concurrently and more importantly to connect.

Raymar LLP example of emergent risk

Lursoft data was key to identifying an emergent risk

Our Head of Financial Crime, Graham Barrow, demonstrates this point of emergent risk in his latest investigation that focusses on Raymar Finance LLP. The company was dissolved in 2018 following a Compulsory Strike Off (CSO) and by referencing a combination of familiar (and some not so familiar) publicly available data points, Entity Investigator flagged high risks in relation to Raymar. In this case, Lursoft data was one of the ‘elements’ that helped an emergent risk to be flagged in combination with other data points.

Increasingly emergent risk interest

Graham explains the case in more detail and highlights data points that would ordinarily be of only mild interest to an investigating analyst (such as changes to designated ownership, dormant account status and links to protected jurisdictions). The interest is ramped up through the identification of repeated overseas investment (Latvia) and involvement with another UK LLP that are the start of building a very different (yet publicly visible) network in which Raymar LLP operates. As always, it is important to recognise that flags generated by Entity Investigator are not necessarily an indication of wrong doing. However, the insight and recommendations available may well prompt a deeper investigation than would otherwise occur.

For more information about Entity Investigator beta testing opportunities, please contact