Using Manual Review Properly 2020
To protect your ecommerce business from fraudsters, you need the right tools for the job. Good fraud prevention tools can identify and screen out fraudulent transactions, but there’s always a balance to be struck. If you reject orders that come with even the smallest red flags, you might reject legitimate customers. If you’re too lenient, the anti-fraud tools can’t serve their purpose. One solution is to manually review the orders that fall into this gray area. Are manual review processes the best way for merchants stop fraud and chargebacks while minimizing the risk of turning away real customers?
When you’re trying to bring down a rising chargeback rate, one of the most persistent challenges is dealing with true fraud. While friendly fraud chargebacks can be fought and won, and merchant error chargebacks can show you where you need to improve your business operations, by the time you see a true fraud chargeback it’s too late to do anything about it. The only thing you can do to beat a true fraud chargeback is prevent the fraud before it occurs.
Tools that screen for fraud during the transaction process can identify many possible fraud indicators: mismatched addresses, suspicious IP or geolocation data, strange ordering patterns—there are all sorts of red flags you can look for, and they can vary depending on the country, industry, even the individual merchant. Anti-fraud screening tools can apply rules that look for these signs and block orders that fit certain criteria, but because of the subjective nature of interpreting certain fraud indicators, they do make mistakes sometimes.
Manual review simply means applying human perception and analysis to fraud-flagged orders to determine whether or not the merchant should accept them.
How Does a Manual Review Process Work?
Manual review can be done in-house, or merchants can outsource it to a third-party provider. When orders that need review are received, they get placed in a queue for the manual review team.
The manual review team will look at each order and examine it for signs of fraud. If the order has already been flagged by an automated anti-fraud system, they will take a closer look at the flagged indicators first.
The team will have a variety of tools and methods to assist them in researching and analyzing orders. At a minimum, they should be able to perform a reverse lookup to verify the customer’s identity and address, and be given access to customer and order databases to review past activity. The team might also be tasked with authenticating orders by calling up the customer or their bank directly.
Ultimately, it is up to the judgment of the manual review team to decide whether the order should be processed or declined. The merchant still gets the final say, but at this point they should be able to feel assurance that every reasonable effort has been made to find out whether the order is fraudulent or not.
What Are the Drawbacks of Manual Review?
While manual review may be a better way to handle orders that skate on the edge of acceptability than simply placing blind trust in automated systems, it’s far from foolproof and comes with a fairly high increase in overhead costs. Manual review can also create bad customer experiences if they take too long to complete.
Small merchants with low order volume and sufficient available staff may not have any issues doing manual review in-house, but the process does not scale up very well. Large, high-volume merchants will likely find themselves looking at outsourcing solutions.
Another issue with manual review, especially if it’s done in-house, is that it relies on subjective human interpretations that can be just as error-prone as rigid rulesets. Each member of your manual review team will develop their own idiosyncratic ways of reading the data and their own personal thresholds for what crosses the line into fraud, and it will take time and experience to develop consistent, reliable judgment. Turnover on the manual review team means starting this process over, which lead to outcomes of varying quality.
When outsourcing manual review, merchants will need to ask some questions up front in order to ensure that the service meets their expectations and delivers a positive ROI, such as how much time manual review will add to the purchase process, and who carries liability for fraud on transactions erroneously approved by the reviewers.
How should I Utilize Manual Review?
Rarely would it make sense for merchants to subject all of their orders to manual review. The best application for manual review is to handle the orders that carry some signs of fraud, but not enough that you feel comfortable rejecting them outright. Again, each merchant will have a different sense of where to draw the line here, and it may take some trial and error to find the balance of acceptances, rejections, and manual reviews that works best for your business.
With this smaller pool of ambiguous orders subject to manual review, you can train and work with your staff or third-party provider to develop the guidelines and methods that produce the fewest unnecessary rejections without increasing your fraud rate.
Merchants should not rely too heavily on manual review as their primary method of detecting and stopping fraud. It’s just too time-consuming to be practical for most merchants, and automated tools can screen out low-effort attempts at fraud much more efficiently.
Almost every merchant has had the experience of trying to make sense of an order with inconsistent or incomplete information and trying to decide if it can be salvaged. Manual review is simply a formalization of this intuitive process that focuses on fraud. For merchants, the question is whether they’re experiencing fraud and fraud-related chargebacks at a high enough rate that it makes sense to invest in a more rigorous and consistent manual review system.
Fraud prevention is key to any chargeback management strategy, and it always requires a multi-pronged approach. Manual review can effectively supplement your automated anti-fraud tools, just make sure you’re stopping enough fraud and recovering enough valid orders that it’s worth the time and labor you’re putting into it.