Holiday Fraud Prevention: Seven tips from the pros
Is anyone ever really ready for the holidays? It’s hard enough to keep up as a shopper, but if you also work in e-commerce fraud prevention, you know you’re guaranteed a crazy ride.
To help retailers navigate the seasonal rush, we crowdsourced some expert strategies from fraud and risk experts who’ve already logged a few seasons in the holiday trenches. Here are their suggestions:
1. Plan for adverse scenarios. Start the holidays with well-prepared SOPs (Standard Operating Procedures) for anything you think could go wrong. Think of things like communication procedures for a fraud attack, handling for staffing or quick decisioning when the manual review queue grows high, and steps to take for internal or external system downtime.
With prepared documentation and agreement from leadership on how to handle fraud risk tolerance in various adverse scenarios, you can focus on fixing issues if they arise instead of figuring out how to fix them.
– Caleb Callahan, director of payments and fraud at Jet.com
2. Adapt fraud logic to seasonal buyer behavior. Anticipate that your customers’ purchasing behaviors may change during the holidays, and update your fraud logic (both your models or rules) accordingly. Depending on your business model, be aware that your average transaction value may increase as well as the count of transactions.
Adjusting your fraud logic will help keep false positives low. You don’t want to find a legitimate user whose order seems to be higher than usual and wrongfully attribute it to fraud!
– Tal Yeshanov, risk and expert, Uber
3. Optimize manual review to enable good buyers. Holidays = abnormal shopping patterns, which can make discerning fraud vs. gift shopping very hard. Since the holiday shopping behavior is abnormal, it can greatly increase the chance that your team will be rejecting good orders incorrectly.
To help mitigate this, find a core two to three indicators of good buyers, then make that fundamental to the manual review process. When your team sees these indicators, accept the order without further analysis. Further analysis will confuse the auditor and increase the chance of making the incorrect decision (most people are risk averse and don’t want to process a potentially bad order). Also, create firm caps as to how long a team should review orders for.
– Courtney Bode, marketplace operations manager at digital mall Wanelo
4. Consider relaxing your fraud prevention. When you are in a good place with your fraud and chargeback levels, you should consider relaxing your protection slightly, to enable more revenue while taking into account a small increase in fraud. This should happen cautiously and in a controlled manner, of course. If you manage this carefully, the increase in revenue will far outweigh the additional fraud risk. The fraud team is also a sales enabler, as much as it is a fraud prevention team.
– Danièle Thillmann, senior VP fraud and customer service at Green Man Gaming
5. Collaborate closely with customer support. Make sure your customer support policies are in line with increased sales from the season rush. For example, if sales increase then contact with your customer support team may rise accordingly. Make sure to work cross-functionally to arm your support team with intel so they can continue to please your customers, which will help avoid chargebacks due to dissatisfied users.
– Tal Yeshanov, Uber
6. Look out for friendly fraud. “Friendly” fraud or unauthorized transactions placed by an acquaintance tend to rise over the holidays – especially as a new wave of younger shoppers start using the computers and devices they just received to shop online.
Keep an eye out for multiple accounts that share a family name, similar geolocation attributes, and payment instruments with a known good user.
– Vinson Lee, business operations at Sift Science, previously trust & safety at Google
7. Keep an eye on seasonal fraud patterns. Make sure someone is watching for trends. Last year, much of our fraud came from Caracas, Venezuela. Once we discovered this, it was easy to reject these orders and narrow down the pool for potential fraud. Once you see trends, be sure to label users in Sift Science.
– Courtney Bode, Marketplace Operations Manager at Wanelo
To help retailers navigate the seasonal rush, we crowdsourced some expert strategies from fraud and risk experts who’ve already logged a few seasons in the holiday trenches. Here are their suggestions:
1. Plan for adverse scenarios. Start the holidays with well-prepared SOPs (Standard Operating Procedures) for anything you think could go wrong. Think of things like communication procedures for a fraud attack, handling for staffing or quick decisioning when the manual review queue grows high, and steps to take for internal or external system downtime.
With prepared documentation and agreement from leadership on how to handle fraud risk tolerance in various adverse scenarios, you can focus on fixing issues if they arise instead of figuring out how to fix them.
– Caleb Callahan, director of payments and fraud at Jet.com
2. Adapt fraud logic to seasonal buyer behavior. Anticipate that your customers’ purchasing behaviors may change during the holidays, and update your fraud logic (both your models or rules) accordingly. Depending on your business model, be aware that your average transaction value may increase as well as the count of transactions.
Adjusting your fraud logic will help keep false positives low. You don’t want to find a legitimate user whose order seems to be higher than usual and wrongfully attribute it to fraud!
– Tal Yeshanov, risk and expert, Uber
3. Optimize manual review to enable good buyers. Holidays = abnormal shopping patterns, which can make discerning fraud vs. gift shopping very hard. Since the holiday shopping behavior is abnormal, it can greatly increase the chance that your team will be rejecting good orders incorrectly.
To help mitigate this, find a core two to three indicators of good buyers, then make that fundamental to the manual review process. When your team sees these indicators, accept the order without further analysis. Further analysis will confuse the auditor and increase the chance of making the incorrect decision (most people are risk averse and don’t want to process a potentially bad order). Also, create firm caps as to how long a team should review orders for.
– Courtney Bode, marketplace operations manager at digital mall Wanelo
4. Consider relaxing your fraud prevention. When you are in a good place with your fraud and chargeback levels, you should consider relaxing your protection slightly, to enable more revenue while taking into account a small increase in fraud. This should happen cautiously and in a controlled manner, of course. If you manage this carefully, the increase in revenue will far outweigh the additional fraud risk. The fraud team is also a sales enabler, as much as it is a fraud prevention team.
– Danièle Thillmann, senior VP fraud and customer service at Green Man Gaming
5. Collaborate closely with customer support. Make sure your customer support policies are in line with increased sales from the season rush. For example, if sales increase then contact with your customer support team may rise accordingly. Make sure to work cross-functionally to arm your support team with intel so they can continue to please your customers, which will help avoid chargebacks due to dissatisfied users.
– Tal Yeshanov, Uber
6. Look out for friendly fraud. “Friendly” fraud or unauthorized transactions placed by an acquaintance tend to rise over the holidays – especially as a new wave of younger shoppers start using the computers and devices they just received to shop online.
Keep an eye out for multiple accounts that share a family name, similar geolocation attributes, and payment instruments with a known good user.
– Vinson Lee, business operations at Sift Science, previously trust & safety at Google
7. Keep an eye on seasonal fraud patterns. Make sure someone is watching for trends. Last year, much of our fraud came from Caracas, Venezuela. Once we discovered this, it was easy to reject these orders and narrow down the pool for potential fraud. Once you see trends, be sure to label users in Sift Science.
– Courtney Bode, Marketplace Operations Manager at Wanelo