With Black Friday and Cyber Monday upon us, swiftly followed by Christmas and the January sales, ecommerce retailers are preparing for their busiest period, and can expect volumes to skyrocket by as much as 30%. Along with an increase in sales however, comes the inevitable hike in fraud.

It truly is a festive season for fraudsters, who take full advantage of the increase in volume and traffic, knowing that retailers’ attention is elsewhere at this busy time.

"By analysing keystrokes, mouse behavior and speed of interactions, it is possible to determine whether it is a human or an automated system"

Additionally this year, US retailers can expect to face an increase in attacks on their online stores, as the roll-out of EMV at the Point-of-Sale will drive more fraudsters online. The temptation is to go one of two ways:

The first is to ramp up your risk settings and treat every transaction as suspicious, but this comes at the cost of conversion rates. Many retailers report a tripling of their risk refusal rate from the industry of average of 3% to 10% during this period.

The second is to accept fraud as an inevitable side effect of this peak period and lower your risk settings. This will lower your false positives but you can expect your fraud rates to soar, leading to lost revenue from charge-backs.

Too much security is almost as bad as none at all

Finding the right balance between blocking fraudsters and optimizing conversions is always a challenge. And during this busy shopping season, a small adjustment in either direction can have a huge impact on the bottom line. The answer lies in a granular approach to fraud defence.

Here are three ways you can target fraudsters while ensuring a frictionless experience for your loyal shoppers:

1. Identify the shopper behind the transaction

Using device fingerprinting and transaction linking you can track shoppers across devices, networks, and online personas and build a holistic view of the behavior of both fraudsters and your loyal customers. In that way you can identify and block fraudsters while allowing your customers to shop unhindered - even if they checkout as a guest or significantly alter their behaviour.

2. Spot suspicious behavior at the checkout

A good way to detect bot attacks is through hyper-specific behavioral tracking during checkout. By analysing keystrokes, mouse behavior and speed of interactions, it is possible to determine whether it is a human or an automated system. In this way you can detect ‘card testing’ and other high-velocity attacks used by fraudsters.

3. Adjust transaction velocity thresholds

Your shoppers’ behavior is likely to change during this period. They might purchase more regularly, or spend more per transaction. To prevent your risk settings from identifying these shoppers as fraudsters, you can keep false positives to a minimum by adjusting your thresholds to anticipate different buyer behaviour.

Brian Dammeir is Senior Product Manager Risk/Fraud at Adyen