Conversion Rate Efficacy and Fraud Reduction Part 2 — Locating the Fraudster
September 8, 2021
Well-researched and effectively implemented fraud management tools detect and hinder future fraudulent activity with efficiency.
If done wrong, however, conversion rates can negatively affect the outcome of fraud prevention systems put in place. For this reason, merchants should adequately evaluate the fraud tools at their disposal, configuring the most salient ones to suit their business and consumer needs.
The wrong combination of tools, or the wrong application process, creates unnecessary barriers for genuine customer (resulting in increased false flags) and likely deters them from shopping with a business in the future. 3-D Secure, for example, is a well-respected fraud management tool, but in certain markets or verticals, causes more harm:
Customers unaccustomed to the tool, such as those in Brazil and China, have higher degrees of poor conversion rates, often resulting in cart abandonment. By comparison, consumer endorsement results in an overall improvement of conversion rates, reassuring customers about the safety of their payments.
Fraud management tools set up based on specific factors that apply to the fraud a merchant experiences (concerning business type, region, fulfillment options and sales channels, among others) is the single most effective deterrent against fraud, as a means of having little or no impact on genuine shoppers.
Indications of Fraud
Traditional identifications of fraud curb incidents of fraud and help accurately determine the legitimacy of transactions. But, as we’ve seen with example like 3-D Secure, each tool needs to be assessed of their own accord and used in a manner that balances sales conversion and fraud detection. There are many available fraud indicator tools that rarely cause issues with conversion rates (device fingerprinting and plausibility checks come to mind). Others, however, require more careful consideration:
Velocity Checks
Velocity is defined as “the number of purchases coming from a specific region.” Merchants can therefore flag transactions if too many originate from a single account, entity, or email address. Certain industries find significant benefits from velocity checks, such as airlines, while others, such as telecommunications and gaming, often suffer from an increased declined of genuine shoppers.
IP Geolocation
A buyer’s location offers key insights into intent (though not without other deciding factors). Comparing a shopper’s IP address against other flagships of identity (BIN range or billing address) help unearth discrepancies. This tool should be taken with precautionary measures, such as manual review versus outright rejections, to ensure there are no negative impacts on conversion rates.
Set Limits
Large purchases mean larger losses if proven fraudulent. Limits allow merchants to set monetary ceilings on the size of purchases. As with the other highlighted tools, context is key. Peak seasonal purchases or new product launches should not face automatic rejection (put other measures in place to protect genuine sales).
Whichever fraud indicator tool is chosen becomes an integral part of the overall fraud management system that relies on consistence and alignment.
New Age Fraud Management Tools & Technologies
Since fraud evolves, so too do the tools used to detect and combat it. The digital age means more sophisticated techniques paired with traditional fraud indicators provide an all-encompassing and effective approach to fraud management.
Machine Learning
A computer’s recognition of patterns applicable to fraudulent and genuine transactions help build algorithms that better predict the likelihood of a purchase being fraudulent. The predictive models help identify patterns far too complex for humans or automated techniques to detect and builds upon itself as it continues learning.
Trained correctly (with the necessary mass of data to do so), machine learning can block fraud behind the scenes with no effect damage to conversion rates. The more accurate the machine becomes, the more it supports stronger conversion rates, reducing false positives and ensuring genuine consumers have a seamless and worry-free transaction process.
Since machines learn from experience, they may “struggle to spot monolithic events and can under-perform when customer buying patterns suddenly move away from the norm.” It is important, therefore, that machine learning be used as a part of an overarching management solution (meant to protect conversion rates and sure up customer loyalty).
Tailored Rules and Alerts
A combination of pattern recognition and flexible fraud guidelines culminate in a dynamic and optimized fraud prevention system whose adaptability enable emerging fraud and buying trends without settling on risk mitigation or costumer experience.
Tailoring fraud rules can help with broad trends: rules flags at the time of fulfillment can trigger 3-D Secure check or introduce additional steps to verify transactions before checkout. Limiting available delivery options may also enable merchants to offer a smooth fulfillment experience for genuine customers.
Silent rule dashboards can offer a clandestine scan of high-risk profiles through which to spot emerging trends, decline orders and avoid future chargebacks.
There are many tools at a merchant’s disposal in the digital age. Picking the right ones for their and their consumer’s needs makes all the difference in the efficacy of business transactions.
In part 3 of the series, we’ll look at the very customers these fraud management systems aim to assist.
Does your small business have a reliable fraud management system? Share how it’s working for you on our LinkedIn page and consider Alto Shield to protect your conversion rates and prevent future chargebacks.
SMEs: Analyzing 2023 And What to Expect in 2024
Mobile Payments: Will They Dominate The Market?
