Skip to content

The true cost of online fraud might surprise you

Join our mailing list

Signifyd regularly publishes free reports packed with business insights, commerce trends and data from our massive Commerce Network. We’ll only email when we have something meaningful to share, no more than once per week. And of course you can unsubscribe any time.

From a  merchant’s perspective, online fraud seems to be as simple as it is devastating.

A bad guy or group of them, use ill-gotten credentials or stolen or manufactured identities to order goods, thereby stealing them from the retailer. And at a high level, that is the long and short of the crime. But the economic effect is much more nuanced and much more damaging.

Signifyd’s Vice President of Marketing Stefan Nandzik sat down last week with EYStudios Meagan Bryson to take a deeper look at the true cost of online fraud and some of the innovative ways retailers, with the help of companies like Signifyd, are embracing new methods to change the math on fraud losses.

First, the problem: Yes, when retailers are hit by fraud, they are out the cost of the goods and additional fees related to shipping and chargebacks. Nandzik explained that cost amounts to about .9 percent of revenue industry-wide. But that’s only a faction of what fraud actually costs, Nandzik told Bryson.

Pilfered merchandise is just the start of online fraud cost

Consider, for instance, a retailer’s cost of steeling itself against fraud — the fraud analysts and detection tools a company buys to spot and prevent fraud. Nandzik puts that at another 1.6 percent of revenue. So, now we’re up to a 2.5 percent bite.

But that’s nothing.

“Really, the biggest impact, when it comes to the cost of fraud, is, as the fraud process matures out, merchants start to get more and more restrictive in the orders they accept, due to the fear of fraud,” Nandzik said during the EY Access podcast.

They start declining orders that are actually good. After all, you can’t receive a chargeback on an order you never send. But overly conservative decision-making accounts for more than half of the cost of fraud, representing 2.9 percent of revenue. All told, Nandzik said, online fraud is costing retailers 5.4 percent of revenue.

And when you apply that cost to the $461 billion in annual ecommerce spending in the United States, you see that such false declines are costing retailers $13.3 billion a year.

Not only is the cost of fraud a big problem, it is growing, Nandzik said. Citing Signifyd’s most recent Ecommerce Fraud Index, he said that fraud losses increased by 7 percent last year.

That said, Bryson asked, what can merchants do to protect themselves from growing fraud attacks?

How to implement effective online fraud prevention

Nandzik explained that some merchants — particularly small and new ones — often begin by monitoring fraud on their own. Some will add a fraud detection in ecommerce tool that provides a score to indicate whether a given order is likely legitimate or fraudulent. But that’s an old model. And one that, frankly, is likely to result in the problem of a merchant declining legitimate orders for the fear of fraud.

There is a better — and newer — way for retailers to protect themselves. Signifyd pioneered the guaranteed fraud protection model, which relies on big data, machine learning and domain expertise. Among the model’s advantages is the vast amount of behavioral and transaction data that feeds the decision-making smart machine that provides a yes-or-no decision on shipping.

“We have a big data approach,” Nandzik said, “tons of digital data, digital body language of people — how does behavior compare between good orders vs. bad orders? We see patterns across sites.”

So, rather than base a decision only on what happens on one retailer’s site, Signifyd’s approach considers millions of transactions on thousands of sites for orders placed in more than 100 countries.

The method also tackles the problems of false declines, by approving the shipment of some orders that appear fraudulent. If the order is indeed fraudulent, that information is immediately fed into the machine-learning model, which hones its accuracy. If the order was not in fact fraudulent, that information also goes directly into the smart machine, which inoculates the system from becoming steadily more conservative.

Balancing fraud fear with customer experience

Best of all, from a merchant’s perspective, the guaranteed fraud protection comes with a financial guarantee. If an order that Signifyd approves ultimately turns out to be fraudulent, Signifyd reimburses the merchant for the cost of the products shipped and any associated costs, such as shipping and fraud fees.

“It’s a balance between fear and the convenience of your customer,” Nandzik said.

The approach elevates fraud protection beyond simply being a security initiative and into being a piece of a retailers’ customer experience strategy. In the age of Amazon, with its fulfillment efficiency, and digitally native retailers, who build a meaningful connection with customers, retailers can’t afford to slow down orders for frequent and cumbersome manual reviews. Nor can they afford to turn down orders from customers who legitimately want to spend money on their sites.

Again, it seems simple: Make buying from your site easy and don’t insult your customers. But like with most things in ecommerce, the concept might be simple, but once you dig in, you see that realizing the vision is not as easy as it looks.

Photo by iStock

Contact Mike Cassidy at [email protected]; follow him on Twitter at @mikecassidy. 


Mike Cassidy

Mike Cassidy

Mike is the head of storytelling at Signifyd. A former journalist and a retail geek, he covers ecommerce and the way technology is transforming digital commerce. Contact him at [email protected].