eCommerce is looking forward to busiest time of the year. In recent years, Black Friday has been the single most profitable retail day in the US and in UK and the trend has now spread Cyber Monday, making it a weekend-long retail festival that acts of as a starting gun for the December Christmas trade. This year Adobe predicts that online shoppers in the UK will spend £24.3billion across the season,[1] 10% more than they did last year, with £901m coming on Black Friday itself.

As the big day approaches, eCommerce merchants have one hundred and one things to prepare. A secure, efficient fraud policy should be at the top of this list. Online retailers that get bogged down in illegitimate transactions or false positives will lose out in the most competitive event of the year. It is important to understand that the threat of overcompensating for fraud is as great as the threat of fraud itself – merchants must ultimately be selective about who they let into their seasonal shopping party. Roberto Valerio, of Risk Ident provides on-point advice on what retailers need to be on the lookout for and how they can avoid this complication.

How do fraudsters strike on Black Friday?

Black Friday not only sees a lot more transactions than normal Fridays, it may also see unpredictable spending patterns. Legitimate customers who rarely make extravagant purchases will often start buying items such as trampolines and television sets, a trend that provides the perfect cover for more sinister activities.

Account takeover fraud is one such sinister activity; a fraudster logs in to a genuine customer account and begins purchasing goods and services without the cardholder’s authorisation. By hijacking the existing account, the fraudster benefits from the genuine customer’s good history and trustworthiness.

The problem is that legitimate customers may display all manner of irregular behaviors on Black Friday. Apart from the obvious change in purchasing habits, customers may also shop at less favored retailers as they chase the best deals (meaning they may easily need to reset long-forgotten passwords) or change their shipping address to direct gifts to friends and families.

What can be done?

One of the more popular defenses is to monitor irregular behaviors such as an unusual number of failed login attempts, a password change followed by unusual behavior or a change of address immediately before ordering.

Ultimately, online retailers can’t rely on rules as they prepare for Black Friday. Machine learning technology, based on developing computer programs, is far more effective. It recognizes patterns and regularities in datasets, and is then able to learn from each transaction and a wealth of historical data. In this way, it can continually create new models and constantly evolving algorithms that allow merchants to drive a more aggressive acceptance strategy while catching fraudulent transactions.

Device fingerprinting will also help trace the routes of fraud. Fraudsters want big rewards fast and will often use the same device for multiple fraudulent transactions. Examining the browser and operating system, language and location, will help identify the source and prevent a Christmas payday for these people.

Artificial intelligence is a vital part of the Black Friday fight against fraud. It’s highly evolved algorithms that can recognize irregularities instantly, accurately picking out the bad transactions even at the busiest shopping peaks.

But merchants shouldn’t depend entirely on technology to see them through – an empowered fraud manager, knowledgeable of the company’s fraud map is indispensable in the process. The best defence is human intelligence allied to machine-level speed and scalability. People commit fraud with the help of technology; we should do the same to counter it.

Roberto Valerio, CEO of software engineering company Risk Ident