The fight against fraud and other types of financial crime has become one of the biggest challenges for global financial organisations. Criminals are using ever evolving methods of attack, exploiting multichannel vulnerability and using massive scale to compromise technology systems. Wire fraud incidents alone have, over the past year, led to million dollar losses at institutions in multiple geographies. Financial institutions must have robust prevention and detection strategies in place in order to protect both themselves and their clients against this threat.
In recent years there has been an explosion in both access channels and transaction volumes meaning that monitoring financial transactions and ensuring that they are secure has become increasingly challenging. The rate of false positives has increased from standard fraud detection rules which leads to more time spent researching and approving transactions.
Big Data analytics are becoming an increasingly essential part of any strategy to help detect and prevent financial crime. Big Data analytics is the current buzz word in technology, but the financial industry has in fact been using real-time analytics for decades, but in todays’ multi-channel world the scale and breath of analytics is growing rapidly.While predictive analytics and variations of other financial crime fighting techniques such as behaviour monitoring, network analysis, pattern recognition and profiling – have been key components of banks’ toolkits for decades. But today, Big Data is changing the game.
While banks have been employing these strategies for many years, Big Data has enabled banks to deploy real-time analytics on a massive scale to meet these growing threats. Financial fraudsters are becoming increasingly sophisticated and daring, leading industry experts to speculate that future attacks could have a systemic impact on the financial system. Financial institutions need to work hard to combat such risks– with those that are unable to protect their customers from cybercrime inevitably risking reputational damage and potential loss of revenues.
To meet this risk, institutions must implement a financial crime risk management strategy that employs a multi-faceted analytic approach to detecting and mitigating financial crime. A range of techniques are used to detect financial crime but the core of any analytics system is built around behavioural profiling. By profiling and tracking the behaviour of an individual account from initial client onboarding, through to transaction monitoring and customer management, it becomes possible to detect unusual account activity.
The stage at which an institutions’ fraud prevention and detection systems will flag a transaction as potentially fraudulent depends on the risk profile of the individual or organisation involved and the defences that financial institutions have in place at any point in time need to be flexible enough to react to evolving typologies of financial crime. Such rapid change requires technologies, models and solutions that can be focused on preventing specific fraud attacks. What’s more, financial institutions must be empowered to effect this change themselves based on their changing risk exposure.
Today’s financial criminals are becoming ever more sophisticated and varied in their attack methodologies. As new forms of payments emerge so, too, do emerging forms of financial crime. A combination of behavioural profiling, real-time detection scenarios and predictive analytics provides the most accurate results. Big Data enables financial institutions to provide these services on a scale that simply was not possible five years.
By Mike Urban, Director of Product Management, Financial Crime Risk Management, Fiserv