TP:  Retailers of all sizes are susceptible to policy abuse though the range and nature of the abuse may vary.

Policy abuse directly affects the business bottom line. Startlingly, studies have shown that 25% of total e-commerce revenue is spent on promotions, returns and refunds! In addition to making payouts for fraudulent claims, businesses waste time investigating and handling claims. Often, by the time the return is processed, the item can no longer be resold to another customer.

The scale of policy abuse is now pushing the merchants to reverse course and change their policies, which could eventually impact sales and their ability to attract new customers and create customer loyalty.

How can organizations balance customer service and customer acquisition priorities with fraud and risk management goals?

TP:  Today, merchants are stuck between a rock and a hard place. Over 90% of merchants rely on promotions, coupon codes and long return windows to attract and keep customers coming back. In a cut-throat retail environment, merchants feel they must continue to offer lenient return policies and promotions to retain customers and stay competitive.

However, policy abuse has made these policies a liability causing significant losses, akin to death by a thousand cuts.

To balance customer acquisition and retention priorities with fraud and risk management, it’s important for retailers to take a holistic and flexible approach. They must customize their customer engagement and policy strategies to align with each market’s unique characteristics, leveraging deep insights and localized tactics.

Additionally, they need to bolster their defenses with sophisticated fraud prevention systems capable of tackling the issue in real-time.

What can organizations do to combat policy abuse? What is the role of AI in combating policy abuse?

TP:  AI and machine learning solutions help make the claims review process less manual and prevent customer representatives from becoming overwhelmed with trying to validate the customer and their history.

Using AI and machine learning, retailers are able to analyze account and transaction data to understand where each instance of abuse originates. Then they can scrutinize customer behavior patterns and highlight only suspicious claims for review, helping to curb abuse while enhancing the customer experience.

Merchants can also enhance their approach to tackling policy abuse by leveraging machine learning clustering technology, which allows them to consider any additional accounts that are connected to a given customer. This helps merchants tackle fraud that arises from the same customer opening multiple accounts, for instance, to use a promotional code more than once.

With advanced AI-driven identity resolution merchants can use a large graph of data and examine dozens of attributes ranging from keyboard language to product type. Also, from the vast range of data merchants can triangulate signals, see relationships among shady transactions, and identify who is behind abusive transactions and what level of risk they pose.

How do you see policy abuse (and mitigation strategies to tackle it) evolve in the future?

TP:  Fraudsters will continue to evolve their policy abuse tactics and retailers must always stay one step ahead. In the future, we are likely to see professional fraudsters leverage more AI and automation to scale their fraudulent activities. Retailers must take steps now to be prepared to fight back and realize their full e-commerce potential.

For instance, Ring (by Amazon) used Riskified’s vast data intelligence to gain clarity into the thousands of cookies, emails, and phone numbers scammers were using to obscure their identities and commit wide-scale fraud. They were able to determine that just 600 individuals were responsible for over US$4m in abuse each year, with some committing as much as $150K annually!

Using Riskified’s Policy Protect automation with Identity Explore technology gave Ring real-time insight that empowered the team to decline orders at checkout that would have led to fraudulent returns. In just seven months, Ring’s fraud team was able to decline $4M+ in abusive returns that would otherwise have represented pure loss.