Large financial institutions deal with many types of fraud, predominantly transaction fraud. Transaction fraud occurs when a fraudster steals credentials, such as a credit card number, and charges those accounts illicitly. To protect against this type of fraud, banks must monitor every card swipe, monitor for unusual or suspicious activity, and then make an immediate decision on whether a purchase is legitimate or fraudulent.
Much like banks, ad tech providers must deal with fraud quickly. Click fraud occurs when a fraudster clicks (manually or automatically) on an advertisement with the intention of inflating click numbers. To detect and deal with click fraud in real time, advertisers need to monitor each click, detect any potentially fraudulent activity, and respond appropriately.
In both cases, the key to success is the ability to monitor, detect, make a decision, and act – all in milliseconds. While post-facto fraud detection has been utilized in the past, today’s customers expect personalized, immediate interactions, requiring firms to move their fraud detection systems from after-the-fact to real-time. This move introduces a new challenge: latency. Customers will only wait for a few seconds when making a purchase before switching to a different card. Thus, the entire purchase needs to be completed in seconds, leaving only milliseconds for fraud detection.
For ad tech providers, new fraud bot networks are successfully imitating both web publishers and users. The fraudsters can spoof popular video content on which publishers sell ad space. Then, bots simulate a human interacting with the video with mouse movements and fake social media information. Detecting and stopping this type of fraud requires a fast system capable of ingesting large flows of both legitimate and fraudulent traffic, and deciding which traffic falls under each category – before authorizing ad spend.
While a third-party anti-fraud solution may seem appealing, outsourcing fraud detection creates a big vulnerability. To be successful, any solution needs to be fast, accurate, and flexible enough to keep up with modern fraud attacks.
To meet the needs of modern fraud detection, companies such as Volt Active Data are offering solutions that can ingest, analyze, and detect fraud hundreds to thousands of times a second. Using stored procedures to manage the analytic logic, paired with state held in Volt Active Data, companies increase their ability to detect fraudulent cards and clicks on the first instance. As blacklists and rules change, these are uploaded into Volt Active Data, where new rules and stored procedures immediately affect processing decisions.
To learn more about how companies such as Volt Active Data are changing how fraud is detected and handled, join us in our webinar on Tuesday, June 13, 2017 at 11:00AM EDT. Our featured speaker, from a leading independent research firm, will discuss how fraud is affecting the financial services and ad tech industries, and how leading financial services and ad tech organizations are fighting back.