The True Cost of Declined Orders

The True Cost of Declined Orders

April 18, 2019

[Editor's note: April is Revenue Recovery Month at Card Not Present. As companies get better at identifying fraud in e-commerce transactions, a danger is that they, their acquirers or the cardholder's issuing bank will decline more legitimate transactions they suspect are fraudulent. Check back here throughout the month for updated content detailing how to minimize declines and maximize sales revenue.]

By Shalhevet Zohar, Riskified

A recent survey on global fraud shows that merchants decline 2.6 percent of all e-commerce orders due to suspected fraud. For orders worth over $100, the rate of transactions declined over fear of fraud is 3.1 percent. For a retailer selling $25 million online annually, this means rejecting orders worth more than $600,000 every year. But the true cost of declines is actually far higher than just the lost sales revenue.

To understand the full cost of declines, one must consider all the factors involved in a holistic risk management approach, including the sales and marketing investment that goes to waste when a good order is declined. Whether online media, SEM, or social media campaigns, e-commerce companies invest considerable funds and energy in customer acquisition and retention—bringing users to the website, bringing them to checkout, and trying to ensure they become repeat customers.

Online travel site Expedia, for example, invested $5.3 billion in marketing in 2017, accounting for more than 50 percent of the entire group's annual revenue. If their decline rate was average, one could say 2.6 percent of that marketing budget, or  around $140 million, was lost due to fear of fraud. Add the lost lifetime revenue from good customers whose order was wrongly rejected, and you start getting an idea of how costly declines can be.

This article describes Riskified’s approach to calculating the true impact declines have on your business.

To fully understand the impact and cost of declines, the following factors should be considered:

  • Customer Acquisition Cost (CAC): What is the cost of bringing users to your site and converting them into customers?
  • Customer Lifetime Value (LTV): How much revenue does the average customer generate for your company over time?
  • Impact of Falsely Declined Orders: What is the financial impact of false positive declines on your business?
  • Impact of Wrongly Blocked Orders: What is the financial impact of blocking orders before they reach your fraud team?

There are plenty of useful resources online to calculate CAC and LTV, both which are crucial to calculating the true impact of declines. But in this article we focus on calculating the value of declined and blocked orders.

Riskified3

Calculating the Impact of Falsely Declined Orders

Some declined orders are indeed fraudulent, and merchants tend to view them more as cost avoidance than lost revenue.

Most merchants estimate 10 percent or fewer of the orders they declined should have been approved. But Riskified’s experience has proven that between 30 and 70 percent of all orders declined by merchants are false declines. This is because most fraud systems and tools are highly risk-averse. Data points that may indicate risk are flagged, and risk management teams must actively strive to approve orders, despite mismatches, in order to avoid false declines.

Below are equations for calculating the financial impact of false positive declines. We’ve filled out the equations using the stats of Merchant X, an average online fashion retailer based in the United States.

Calculating the Impact of Wrongly Blocked Orders

It is often the case that some transactions are rejected or filtered out before even reaching the fraud team. For instance, many merchants deploy systems that block orders from “risky” geographic locations, because they feel accepting orders from these locations is not worth the risk. The decision to reject all orders placed from certain countries is sometimes based on bad experience—a significant sum was lost to a chargeback, for example. But Riskified research has found that even orders from the riskiest countries are still mostly safe, and simply blocking these orders is sure to leave money on the table.

Another common reason for rejecting orders on the gateway level is AVS (Address Verification System). Payment processors encourage merchants to set automatic AVS mismatch filters as an anti-fraud measure. A large online fashion merchant, for example, was using AVS filters before moving to Riskified. Twenty percent of the orders Riskified approves for this merchant would have been rejected by their AVS filters (read more about AVS mismatches and their impact on fraud).

3-D Secure is yet another service that can lead merchants to inadvertently reject orders from legitimate customers. While this payer authentication service doesn’t block orders per se, it adds additional verification steps to the checkout process, impacting not only customer experience but also sales conversion rates. In other words, utilizing 3D Secure effectively results in a high drop-off rate, preventing many legitimate orders from even making it through to the risk management team. An update to this process—3-D Secure 2.0—is currently being rolled out, but even this solution will introduce some friction.

To sum up, in addition to the value of the orders that are falsely rejected before reaching your fraud team, one should consider the potential lifetime value that is lost when a legitimate customer’s order is wrongly blocked.

The following equation can be used to calculate the financial impact of legitimate, blocked orders:

Riskified4

Correctly calculating the financial impact of wrongly blocked orders can be tricky. First of all, merchants do not always have the required data. Many online stores don’t know how many incoming transactions are blocked by rules or filters such as those mentioned above. If you don’t know all the metrics, ask your payment processor for these numbers.

Second, using the same customer LTV for blocked orders may be inaccurate. Riskified’s data shows that cross-border orders are actually worth far more than those placed domestically. Orders placed in the U.S. with cards issued in Latin America or the Middle East, for instance, are worth 41 percent more than those placed with domestic U.S. cards. If you are rejecting international orders or orders placed with cards that don’t support AVS (anywhere outside the U.S., U.K. and Canada), you are likely turning away some very high-value customers.

Lastly, calculating the true rate of falsely blocked orders is very hard. The insult rate, or number of customers who contact you complaining over a rejected order, is much smaller than the actual amount of good customers you’re turning away.

The End of Declines Starts Today

The financial impact of blocking and declining transactions is largely underestimated by e-commerce merchants. Reducing the amount of legitimate orders that are rejected can have a significant impact on the company’s overall performance. In Riskified’s experience, removing filters and reducing false positives leads to significant increase in merchants’ sales revenue. Therefore, taking the time to calculate the impact of declines on your business and working to reduce false positives is well worth the effort.

Previous-Article-CNP Next-Article-CNP

Setting Standards to Prevent, Detect and Respond to CNP Fraud

  • Share this Article:
Shalhevet Zohar