What is Your Data Worth in the Fight Against Fraud?

You don’t need us to tell you that fraud and financial crime is on the rise. A quick google search will give you endless stats to support this claim. Fraud losses are increasing as a percentage of revenue, and that direct impact on the bottom line isan area of laser focus for senior execs. 

The theme of this blog is about what your data is worth. Let’s try and put that into context. According to LexisNexis, the level of fraud as a percentage of revenues has increased, from 0.95% in 2017 to 1.53%, on average, in 2018. That means for any reasonably sized organization, the worth of your fraud related data sources is probably in the region of tens if not hundreds of millions of dollars per year. Furthermore, data is the primary raw material used to understand and detect fraud.

Great — now that I have your attention...

New technology and services such as contactless payments make fraud easier, at the cost of convenience for the masses. Is this a price worth paying? Maybe, but fraud comes off the bottom line, and with many financial services firms working to tight margins, it becomes an obvious place to focus on. Not only does preventing fraud improve profits, perhaps more importantly it improves brand reputation too.

Fraudsters are concocting complex strategies and scams to defraud innocent people, by exploiting the gaps between people, process and technology. Take for example, the silo that typically exists between security, fraud and compliance teams and their disparate tooling, reporting lines and datasets.

Firms are looking more closely at how they can use their data to detect more fraud sooner.

Existing fraud tools on the market ARE data-driven but have not prevented the increase in fraud. This is in part due to two factors:

• Fraud models are typically being built on structured data and miss insights gained from unstructured sources.
  
  
• Point fraud solutions address individual areas of fraud and are unaware of moderately suspicious behaviour across multiple sources that may amount to serious or complex fraud.
  
  

Analysing and modelling fraud with structured data alone doesn’t go far enough because it doesn’t encompass all of the interaction points and channels a fraudster has with an organisation, leaving blind spots in detection logic and monitoring. Only logs provide insights into website, call centre and mobile app activity, critical in building a behavioural understanding of how complex fraud takes place, starting with account take-over attempts through to money laundering and financial crimes.

Websites and apps are the shop window of today and fraud teams should have a full view of activity so they can understand the attack vectors and scams leveraged by fraudsters and put effective controls in place.

Companies can gain more value from their data through improved correlation across structured and unstructured sources by leveraging a data platform like Splunk. Splunk can handle the challenges of unstructured data which typically arrives in high volumes, at high velocity and in many varieties and formats. These challenges are the reason why fraud teams tend to avoid modelling these types of data sources; however, Splunk makes it much easier to consume these data sources and model for fraud in combination with structured sources to aid the detection of more fraud.

By combining logs and transactional data, organisations can identify fraudulent behaviour that would otherwise go unnoticed or detect it sooner to minimize the fraud. 

Splunk can become the backbone of an anti-financial crime operation so that you too realise more value from your data.

----------------------------------------------------
Thanks!
Haider Al-Seaidy