Rafael Lourenco
When you’re doing business online, it can be hard to know if your fraud prevention solution is doing what you need it to.
That’s why understanding key performance indicators (KPIs) for fraud is critical for e-commerce merchants.
Knowing and monitoring these KPIs will help you identify potentially fraudulent transactions, increase customer satisfaction and build revenue.
Regularly tracking these 13 fraud KPIs will help you stay ahead of potential fraud issues, preventing small problems from developing into potentially devastating crises that can shutter your business.
1. Order Approval Rates
When it comes to minimizing fraud, you need to know what percentage of your incoming transactions are ultimately approved. But you might not be calculating your order approval rates correctly.
You may be omitting auto-declines from this calculation, falsely assuming these orders weren’t legitimate to begin with. But this omission can result in you artificially inflating your actual order approval rate.
For an accurate order approval rate, include all incoming transactions into your calculation.
2. Chargeback Rates
If your chargeback rates trend above 1% of revenue, you may find yourself labeled a high-risk merchant (at best) or may even have your merchant account cancelled completely (at worst), leaving you unable to accept credit cards for incoming orders.
Adding to the stress, chargeback rates (the number of chargebacks you receive in a fixed period divided by the total number of sales for that same period) are calculated differently, depending on the credit card processor. For example, Visa uses the total sales for the current month; MasterCard uses the total sales from the previous month.
But to get the most accurate view of what happened in a particular sales month, you should divide the chargeback loss amount by the sales number from the month in which the transaction that generated the chargeback was originally placed. For example, if the original transaction took place in May and the chargeback took place in June, you would divide the chargeback loss amount by the sales number from May.
You might even consider analyzing your chargeback rates further by breaking the rate down by individual payment methods. You may find your e-commerce store is more vulnerable to fraudulent transactions using alternative payment methods, like PayPal or ApplePay.
3. Manual Review Rates
High numbers of manual reviews can take a toll on your business. Rising numbers mean you’re spending an increasing amount of time, money and resources evaluating the legitimacy of a transaction. But too-low rates could indicate you’re automatically declining legitimate orders.
4. Average Time for Order Reviews
Customers don’t like to wait when shopping online. They want to get on your site, find the item they need and complete the transaction. But when orders are pending approval for too long, it could indicate that you have a problem with your fraud prevention program. Most orders shouldn’t have to go through an extended manual review process, and long review times should be the exception, not the rule.
5. Checkout Abandonment Rates
With checkout abandonment costing merchants an estimated 70% of order value, you need to be calculating your checkout abandonment rate, which is the number of orders completed divided by the number of checkouts initiated. Too-high rates suggest there’s a problem with your checkout experience.
This increased friction might be due to fraud prevention features like 3D Secure and secure customer authentication on your website. If this happens, you’ll want to re-evaluate your checkout process and see how it can be streamlined for legitimate customers without offering entry points for fraudsters.
6. Automatic Decline Rates
You might be so concerned about fraud that you’ve been tightening up your fraud controls, hoping that will reduce your fraud exposure. But as your automatic decline rate (the number of completed or attempted transactions that are automatically declined, divided by the total number of orders) increases, it could be a signal that you’re actually auto-declining a larger percentage of “gray area” transactions —many of which could actually be good orders.
7. Total Decline Rates
Total decline rates — the total number of orders you decline (including auto-declines) divided by the total number of orders — are one of the most critical KPIs for determining how big a problem fraud is in your business.
8. False Decline Rates
No merchant wants to decline good orders, but it happens. But knowing your false decline rate, or how many good customers are impacted to capture a single instance of fraud, can offer great insight as to the effectiveness of your fraud prevention strategy. Bigger numbers mean you’ve declined more good transactions, and that results in more than just lost sales. It creates poor customer experiences and loss of the lifetime value of the customer. Unfortunately, this rate is notoriously hard to calculate, and you may need to resort to monitoring churn rates, tracking complaints received by your customer service department and auditing a random sample of the declines to get a better handle on your total false declines.
9. Average Confidence Rating of Order Decisions
Subjectively assessing your confidence in an order decision and assigning a value to it may seem to be an arbitrary way to measure fraud, but it can actually be helpful. Transactions may be marked as “high confidence” orders when the customer confirms the order, while “lower confidence” orders may be ones that your fraud filter auto-declined. If you see too many “low confidence” orders coming through your store, you might find you also have a problem with fraud.
10. Cost Per Analysis
If you want to know how cost-effective your fraud prevention system is, start calculating this KPI. To get the most accurate number, you’ll want to include all expenses related to manual and automatic analyses (including employee salaries and overhead) and the lost customer lifetime values from falsely declining orders.
11. Fraud to Sales Ratio
When comparing the ratio of your dollar amounts lost to fraud compared with those amounts gained through sales, you need that number to be as low as possible. If you notice the ratio starting to rise, your fraud prevention solution probably is no longer doing its job.
12. Account and Value Detection Rates
These two rates measure the percentage of correctly identified fraud accounts (account detection rate) and the amount of money saved because of that correct identification (value detection rate). These numbers can help you identify the volume of fraud that escaped undetected, and they can also help you objectively measure the value your fraud prevention system is offering.
13. ROI of Fraud Tool
This KPI, like the Value Detection Rate, helps establish the value of your fraud prevention system, demonstrating the net benefit to your business of the fraud tools you have in place. To calculate this KPI, you’ll need to know several figures, including your fraud dollars saved, your sales dollars lost and the cost of your daily operations.
Too-low ROIs suggest that your fraud prevention strategy isn’t doing all it can to identify and block fraudulent transactions or that the cost of the tool is simply too high for what it’s delivering to your business.
Getting Started With Fraud KPI Tracking
To make smart business decisions, you need to be tracking the fraud KPIs that are most relevant to you. At ClearSale, we think this is so important that we make reviewing fraud KPIs part of the onboarding process with every new client. This, combined with our broad experience across all industries and global markets, lets us accurately analyze your true risk profile.
Our goal is to help you understand what your fraud prevention solution is doing well, where it falls short and what the ClearSale solution can offer your business. Contact one of our fraud prevention specialists to learn more about our approach and determine if we’re the best solution for your growing business.
