The Simility Blog

Do You Have a Legacy Fraud Solution? Here Are Eight Tell-Tale Signs
Jayan Tharayil
October 16, 2019

It’s no secret that fraud is on the rise, fueled by persistent data breaches, and the widespread use of anonymizing technology and automated account takeover tools. Over 14 million consumers fell victim last year, and this estimate by Javelin Strategy & Research could be just the tip of the iceberg.1 Against the growing volume and sophistication of fraud attacks, organizations armed with legacy solutions are struggling.  

Without the right tools in hand organizations threaten to be overwhelmed by chargeback costs, administrative overheads, regulatory charges and customer churn. That’s partly why many US financial institutions plan to upgrade or replace their fraud solutions over the next 1-2 years, according to Aite.2

With this in mind, we’ve put together eight tell-tale signs that your current fraud prevention solution needs replacing:

1. Your Fraud Analysts Are Overloaded 

The best data-centric tools should be able to automate most fraud decisioning, leaving just a handful of cases for manual review. This reduces overheads and improves the customer experience. If your analysts never have a spare second, those tools aren’t performing as they should. 

2. Your Customers Are Abandoning Their Interactions 

If customers are walking away in large numbers, there’s clearly something wrong. It is most likely the result of tools which are unable to make intelligent, risk-based decisions around each transaction. That means they may be flagging large volumes up for manual review (see above) or initiating additional, unnecessary authentication checks. Both add extra friction for the customer which can put them off interacting with your business altogether. 

3. Your Solution Only Detects Known Patterns of Fraud

Fraud is in a continual state of flux. Scammers are in constant conversation with each other on forums and dark web sites, sharing the latest tools, techniques and expertise to outwit fraud prevention platforms. Tools which only detect known patterns of fraud may be fine for a while, but what happens when, inevitably, those patterns change? You need to be highly adaptable. 

4. You Don’t Know Why Some Transactions/Events Are Declined, While Others Are Not 

To tackle fraud successfully despite this constant evolution of black hat techniques, you need to not only automate and improve decision making, but also improve transparency. Without knowing why some transactions or events are declined, there’s no way to continue evolving your risk management strategy. 

5. You Can’t Combine and Analyze A Variety of Disparate Data Inputs And It’s Difficult To Add New Feeds 

Effective fraud prevention ultimately boils down to the quality of data you use, and how you use it. That means leveraging a wide variety of data sources. These could include device information, network intelligence, session and behavioral information, and much more. There’s no definitive list here and important new sources may emerge over time which you also need to include. 

6. It’s Difficult To Add, Configure, and Modify Rules and Machine Learning Models 

As mentioned above, the next crucial step is how you use and interpret that data. Because fraud patterns are constantly evolving, you’ll need a solution which is extremely adaptive and agile, and capable of a high degree of customization. Automation and machine learning are great here, but they can’t replace the skills of your in-house team. That’s why it’s vital you find tools which enable them to write, modify and implement rules manually in plain English, as well as machine learning models that can react, adapt and evolve over time independently as fraud changes. 

7. You Can’t Test Rules and Models For Efficacy to Understand The Impact of Changes Before Implementing Them 

A vital part of the process for adding and updating new rules and models is testing. But many tools either don’t allow for comprehensive testing or introduce unnecessary delays which only give fraudsters the upper hand. The best advanced platforms will allow testing within minutes for fast, adaptive fraud prevention. 

8. You Can’t Easily Configure Your Decision Orchestration Layer to Automatically Approve, Decline, Step-up Or Review Based On Your Own Business Needs 

Making the right decisions is crucial to effective fraud prevention. But doing so in an automated way is also key. Otherwise your fraud analysts will be overwhelmed and the customer experience will suffer. But every organization is different. That’s why you need modern decisioning platforms in which the orchestration layer can be easily customized according to your risk tolerance. A simple rule of thumb is that low value, high volume transactions are scored lower risk and are more likely to be approved without extra authentication checks while low volume, high value payments represent a greater risk and could require two-factor authentication or manual review.  

Configuration and automation are key to maintaining that all-important balance between minimizing fraud and optimizing the customer experience. To learn more about how Simility’s industry recognized Adaptive Decisioning Platform can help improve your fraud prevention strategy, schedule a demo now.

1 Javelin, 

2 Aite,