The Lifecycle Of Fraud — A Dynamic Solution

By PJ Rohall, Supervisor, Fraud Management Solutions

It is scary to think some fraud prevention methods still rely on personally identifiable information (PII) as the sole means of combating e-commerce fraud. In a world riddled with data breaches, this form of static identity assessment and authentication is not only antiquated, but will soon be extinct.

Let’s face it, PII is something criminals have and screening for it alone will leave your company exposed. Not only should companies rely on dynamic data, they should also implement an all-around dynamic solution when battling fraud in the card not present space. The same way dynamic data strengthens fraud prevention — pulling from multiple industries, channels, and devices — a dynamic solution provides a more strategic and agile method of fraud management. It spans the entire lifecycle of fraud, addressing every stage with an ecosystem of people, processes, and technology. A guided tour through the lifecycle of fraud will unlock the value of a dynamic solution.

Data breaches have provided a seemingly endless supply of personal and financial information to the criminal world. Armed with this sensitive information, criminals take to their mobile devices, tablets, or computers to cash in on free merchandise. Once the criminal clicks submit, a dynamic fraud solution unleashes its first line of defense in the form of a fraud technology lab. Here, data scientists harvest dynamic data in order to optimize antifraud technology, utilizing machine learning and predictive analytics.

Cutting edge technology, paired with brilliant analysis, helps to lay the foundation for the overall solution. This complex analysis produces exceptional intelligence, helping to feed a robust rules engine. A rules engine should be incredibly efficient, holding an extremely low percentage of orders for review enabling the majority of good orders pass through. But what makes the rules engine dynamic is its ability to scale.

It should be able to flex up or down, adjusting to increases in volume and shifting fraud patterns. Holiday volume, sales, and fraud trends can wreak havoc on merchant operations, leaving customer satisfaction hanging in the balance. A rules engine should adapt to these demands, mitigating fraud and maximizing conversions regardless of the volume or type of fraud it encounters.

When a fraud order is held for review, it collides with a well-trained fraud investigation team. These investigators are dedicated to piercing fraud detection and are measured on quality, efficiency, and customer satisfaction. The last KPI may surprise you, but the best fraud investigation teams place a considerable emphasis on converting orders, driving sales, and providing a tremendous customer experience. The review process as a whole should be customer centric, prioritizing orders based on the customer’s expectations on receipt of goods or services, utilizing features like order age and delivery method.

Another layer of review performs macro-analysis on large data sets in order to identify pockets of fraudulent orders (as well as good ones). Think of this as taking a look at the forest, instead of each individual tree. With tremendous efficiency, these data analysis gurus perform high level linking and tagging. This enriches data quality and helps adjust to sophisticated fraud attacks. The intel is disseminated quickly, allowing all teams to absorb the changing fraud landscape and adjust their risking procedures as needed.

What if a fraudulent order evades the initial layers of detection, and is fulfilled and loaded on a truck? Even then, it is not too late. Performing post transactional analysis enables analysts to identify fraudulent transactions and send stop shipment requests. Fraudulent orders can be stopped prior to shipping or while in transit, leaving a confident criminal empty handed. This analysis requires seamless interaction between a fraud department and order fulfillment specialists. The quicker and more agile the solution, the more likely fraudulent transactions are successfully upended.

However, some lucky criminals will receive their merchandise. While frustrating, these orders are gone but never forgotten. Criminals leave behind a trail of data that is extremely valuable to a fraud management team. Like blood returning to the heart, this data flows back to the fraud technology lab where data scientists extract actionable intelligence. Model analysis, rule analysis, and chargeback analysis are the vehicles that re-oxygenate the solution. The strategy and analytics team and fraud investigation team collaborate in order to perform the most thorough analysis. Each team has a unique perspective on the data and together they provide tremendous intelligence. Furthermore, data scientists utilize predictive analytics to ensure each model is operating at maximum efficiency. This data crunch is a collaborative effort across multiple teams with each team dialed into the role they play in order to spin a web of well-coordinated intelligence.

Once an order is disputed as fraud, the money is lost — not necessarily. A chargeback recovery team is implemented with the sole purpose of recovering lost funds. This team is particularly important for ecommerce merchants as they are liable for fraudulent transactions. The chargeback recovery team implements an aggressive and strategic methodology to fully vet and fight fraud disputes. They take advantage of issuer policies while supplying ample evidence in order to win cases and recover funds. Credit card issuers literally wrote the book on chargeback rules and procedures, therefore merchants who are uneducated and unprepared will be at a severe disadvantage. Make sure you have a team of experts to fight for every dollar and leave no money on the table.

Dynamic data is a critical component when competing in ecommerce fraud management. However, you cannot stop there. A fraud solution should blend human ingenuity, powerful technology, and innovative processes. These components must be utilized and layered in a manner that fosters dynamic intelligence. By spanning the complete lifecycle of fraud, this agile solution will alleviate merchant pain points, meet customer expectations and adapt to the evolving ecommerce fraud landscape.

PJ Rohall is a Supervisor on the Fraud Management Solutions Team at Radial. He has eight years of professional experience, spanning Financial Services and Consulting Organizations, most recently joining Radial’s Fraud Team. PJ manages a team of Fraud Analysts for their best-of-breed solution. His team disrupts conventional fraud detection by driving order conversion and balancing fraud mitigation with delivering a tremendous customer experience. Immersed in card not present fraud management for the past 4+ years, he specializes in generating solution awareness and executing strategic marketing campaigns. PJ holds a Bachelor’s Degree in Business Administration from The University of Richmond.