It can be difficult for merchants to really understand how well they are performing with manual reviews for online fraud management relative to their peers. In this FraudBlog post we look at merchant reported statistics from three surveys covering merchants of all sizes in North America and the United Kingdom highlighting major differences between high and low revenue e-tailers.
Before delving into the statistics, it must be noted that every merchant is different. The appropriate number of manual reviews performed, required staff as well as acceptable loss and conversion rates are contingent on volume, margins, fraud loss amounts and other merchant-specific factors. However, there are some trends that persist across merchants with similarities in volume or types of goods or services sold online, and such merchant reported statistics can be helpful for organizations who want to assess their manual review performance against what others are reporting.
For tips on quantifying the costs and performance of internal manual review staff and processes check out this recent white paper from The Fraud Practice. Although once a merchant calculates these metrics and estimates their total costs, it is helpful to compare this against what others are reporting in the industry.
One key metric is the manual review rate, or number of transaction attempts that are queued for a manual review. This varies greatly by merchant volume as merchants processing tens or hundreds of million transactions per year cannot feasibly review a high portion of order attempts. Many start-up and smaller merchants, on the other hand, may review nearly any transaction that isn’t squeaky clean.
A relatively recent survey from Internet Retailer found that just over 40 percent of merchants review 5 percent or less of all online order attempts while 53 percent of merchants review no more than 10 percent of order attempts. A little more than 7 percent of merchants indicated they manually review between 11 and 20 percent of all order attempts, another 7 percent review between 21 and 40 percent of orders and 3.5 percent of merchants review 40 percent or more of transactions. Based on this distribution heavy towards the lower end it is likely that many of the merchant survey respondents were large or enterprise organizations in this study.
Studies from CyberSource provided more detail and insight to average manual review rates sliced by merchant revenue. According to their most recent North American merchant survey 27 percent of online orders are reviewed overall, but this varies greatly depending on merchant volume. Online retailers grossing more than $100 million per year online manually review 7 percent of orders on average compared to a 40 percent manual review rate on average for merchants grossing less than $5 million per year in online sales. Merchants selling between $5 million and $100 million worth of goods and services online per year manually review about one in every four orders.
Figures from CyberSource’s UK merchant survey are comparable across all merchants overall with an average review rate of about 25 percent. However, the largest merchant revenue group (> £25 million) tends to manually review more transactions relative to North America at 16 percent, more than double the review rate for high online revenue merchants. UK merchants in the lowest online revenue grouping (> £500,000) review closer to one-third of orders compared to 40 percent for North American merchants.
Merchants should monitor manual review rates to be sure that there aren’t too many, or too few, orders being queued for review. One way to check on the health of this is to look at the manual review accept and decline rates. This refers to the outcome of the review (accept/deny) relative to the total number of orders reviewed. If a very high number of reviewed orders are accepted, then chances are there are things the merchant can do to reduce the number of good orders going to a review agent.
According to the North American CyberSource merchant survey the average manual review accept rate across all merchants in 85 percent. Merchants surveyed in the UK, however, reported a manual review accept rate of 71 percent. Of course the manual review accept rate needs to be looked at in context of the fraud rate on manually reviewed orders. If there is a high manual review accept rate but a high portion of these reviewed orders are accepted and end up as fraud, then the problem is more likely related to the quality of reviews than the quality of orders being queued for review. Only two-thirds of the merchants CyberSource surveyed in North America tracked fraud rates on manually reviewed orders, and these merchants reported a missed fraud rate of 3 percent on orders they sent to review.
Merchants should also consider how much time they are spending per review to help estimate the costs of labor. North American merchants reported in the CyberSource survey that order review staff alone represent more than half of all fraud management expenditures. Merchants selling more than $100 million worth of goods or services online spend an average of 4 minutes per review in North America, compared to 10 minutes per review for merchants with annual online revenues below $5 million.
If a merchants does business in the online channel it is likely that they are performing manual reviews. About 81 percent of merchants in North America and 58 percent of merchants in the UK conduct manual reviews according to the two CyberSource surveys. Even a merchant believes their manual review processes and performance are healthy, it is a good exercise to assess the costs and production of manual reviews and compare this against what others have reported.