The eCommerce Guide to Passing More Medium and High-Risk Orders

Legitimate Customers Get Denied

Strict risk criteria may flag perfectly legitimate transactions as high-risk due to minor discrepancies or unusual behavior. This can lead to loyal customers having their transactions declined, causing frustration and potentially driving them away.

Lost Revenue

False declines can result in lost sales and revenue for businesses. Customers who experience a declined transaction may abandon their purchase altogether or seek out a competitor with a more user-friendly payment process.

Customer Dissatisfaction

False declines create a poor customer experience. Customers encountering this issue may be less likely to return to the same merchant or recommend it to others.

Reputation Damage

Repeated false declines can damage a merchant’s reputation. Customers who consistently experience difficulties with their payments may share their negative experiences online, harming the brand’s image.

High Maintenance Costs

Depending on stringent rules alone to manage risk requires constantly monitoring and changing those rules as fraud trends, shopper behavior, and your business continually change. This adds unnecessary overhead costs to your team, without additional value, and will likely still result in false declines because of the inherent inflexibility of this approach.

Travel Spending Patterns

Cardholders often don’t inform their bank or credit union of travel plans, causing their
purchase behavior to be flagged as unusual. When customers make purchases from
locations significantly different from their usual geographic area, it can trigger fraud
alerts. This is especially true for international travel.

Moreover, when travelers have trouble with payment, they may make multiple attempts to complete transactions. And as they’re traveling, they’ll make frequent transactions from different cities or countries within a short time frame. All of these actions taken during travel can be seen as unusual and suspicious, leading to more false declines.

Large Purchases

Shoppers make big-ticket purchases less often; and when they do, it can appear outside of their normal spending pattern. Payment processors and banks often employ fraud detection systems that flag large transactions as potentially high-risk. While these systems are essential for preventing fraud, they can occasionally result in legitimate transactions being declined.

Gifting

Shoppers often change the shipping information on multiple orders when sending out gifts. When customers make multiple order attempts in a short period, it can trigger velocity checks designed to detect potential fraud. This is because fraudsters often try multiple transactions in quick succession to exploit vulnerabilities. However, legitimate customers may place multiple gift orders during holiday seasons or special occasions.

IP & Billing Address Mismatch

Transactions made from locations that don’t match the customer’s billing address can trigger false declines, especially if the customer is traveling or using a VPN. A mismatch in IP address and billing address is also very common for business purchases as the shopper is completing the purchase for a business that can be located in a different country.

AVS Mismatch

When there is an AVS mismatch, it means that the address entered by the customer doesn’t match the address on record with the credit card issuer. While AVS checks are designed to prevent fraud, mismatches often occur due to a typographical error or the address on file being outdated. If a customer accidentally enters the wrong billing address, even if it’s a minor mistake, the payment processor may flag it as a potential fraudulent transaction and decline the payment.

Even worse, clunky checkout experiences can confuse shoppers. When merchants ask for the shipping address first, it can be unclear to the shopper where to input the billing information — which is commonly a tiny checkbox that is easily overlooked.

It’s also important to note that some payment processors don’t support P.O. Box addresses; so if a customer provides a P.O. Box as their billing address, it might result in a mismatch. And since AVS is primarily designed for U.S. addresses, it may be less effective for international transactions and can result in more frequent mismatches.

Billing & Shipping Address Mismatch

Having different billing and shipping addresses is not necessarily an indication of fraud. Many legitimate reasons exist for shipping to a different address, such as the recipient not being available during the day to accept a delivery, sending a gift to a friend or family member, or even using a workplace address for delivery.

Outdated Payment Information

In some cases, shoppers may not be aware that their card issuer has issued them a new card with updated details like a new expiration date or card number. When they attempt to use the old card details, it can lead to declines.

Don’t over-rely on automation.

Flagging orders based solely on a single anomaly like an AVS mismatch has proven to be more of a revenue blocker than an effective fraud prevention tactic. To safeguard your shop while also capturing more revenue, be sure to consider all order characteristics to determine the accurate risk of the transaction. Too many merchants will have very stringent rules to block orders where the billing and shipping don’t match or if the AVS is not a match, resulting in an increase in false positives and uncaptured revenue.

Additionally, it’s important to regularly review and fine-tune risk criteria to adapt to evolving fraud tactics and changing shopper behavior. A fraud prevention platform isn’t meant to be a static solution, but rather it’s dynamic and built to evolve and adapt to new shopping — and fraud — landscapes.

Recalibrate velocity check triggers.

When customers make multiple order attempts in a short period, it can trigger velocity checks designed to detect potential fraud. This is because fraudsters often try multiple transactions in quick succession to exploit vulnerabilities. However, legitimate customers may place multiple gift orders during holiday seasons or special occasions.

An effective fraud prevention solution can differentiate between genuine customers placing multiple gift orders and potential fraudsters. Work with your fraud prevention provider to implement more sophisticated fraud detection algorithms that analyze customer behavior over time which will help distinguish between these scenarios. The fraud prevention solution can set up customized fraud rules and thresholds that consider the unique characteristics of gift-related transactions. For example, they can set rules that allow for multiple shipping address changes within a single purchase session or adjust velocity thresholds during peak gift-giving seasons.

Implement a proactive review process.

Determining whether a transaction is trustworthy or not is usually quite straightforward. Most fraud detection software uses algorithms that analyze data from numerous sources and decide whether there are enough trust factors or risk factors to approve or decline an order

However, the challenge arises when dealing with transactions that fall in the middle of the risk spectrum. These are transactions that do not display obvious indications of trustworthiness or fraudulence, or which may have signs of both. NoFraud refers to these transactions as “grey” orders, as they contain a high degree of uncertainty, with a significant number of both trust and risk indicators.

For example, let’s say an order shows some suspicious signs: the address listed with the bank doesn’t match the order, and the email used to place the order is brand new. These could be signs of a fraudster at work, or they could be signs of someone who just started a job in a different city making a purchase using their new company email.

Since this transaction also shows trust factors (CVV match, no risky device detected, etc.), it would likely fall under the category of a “grey” order. So without further review or verification steps, the determination of whether this order gets a pass or a fail is going to be dependent on your fraud prevention provider’s risk tolerance.

Rather than over-relying on automation, all “grey” orders should automatically be escalated to a skilled review team — preferably one that is reviewing transactions 24/7. If you’re using a solution like No Fraud, a proactive review team will also review select high-risk orders, such as orders with a very high order value, in order to capture more revenue.

In the example of the suspicious order above, a NoFraud expert fraud analyst would determine through additional research that the purchaser has started a new job at a company whose name matches the email used for the order. Alternatively, with your permission, they could request documentation to verify identity, giving the transaction a higher chance of being approved. In this case, after review, our analysts would pass the order that other fraud software would automatically decline.

Give shoppers the opportunity to correct typos before failing the transaction.

When customers modify shipping information or enter an incorrect address, it can trigger fraud detection systems to flag these transactions as potentially suspicious. This is because fraudsters may attempt to change shipping details to obscure their activities. However, legitimate customers may also need to update shipping addresses for genuine reasons, such as sending gifts to different recipients or updating an old address.

Maximize your approval rates by implementing a pre-gateway fraud prevention system that operates in conjunction with a dynamic checkout process. This checkout process will prompt customers to correct any typos or incorrect information in real-time, rather than screening orders post-gateway when it is too late to recover any orders that have been rejected.

Improve risk criteria by analyzing data from third-party sources.

Rejecting an order based on one or two factors is detrimental to business. It’s important to regularly review and assess risk criteria as fraud attacks and shopping behaviors continue to evolve. The data on the order can tell you a lot but it helps to utilize other sources to obtain additional insight into each data point. By analyzing third-party data, merchants can use the additional insights to make more informed decisions about the legitimacy of transactions.

Beyond shipping and address verification or geolocation data, implement device fingerprinting solutions to analyze the device used to place the order. Compare device characteristics like IP address, browser type, or device type, with known patterns of legitimate customer behavior and known fraudulent activity. Be sure to incorporate third-party data sources into your fraud prevention solution’s machine-learning models, so it so it can continuously learn from new data and adapt to emerging fraud trends — ultimately helping to improve predictive accuracy. The best fraud prevention providers are looking at thousands of data points — not a handful or even hundreds — in real-time to determine the legitimacy of a transaction.