Webinar Recap: How to Win at Every Stage of the Chargeback Cycle
April 13, 2017 -- If you missed yesterday’s webinar “How to Win at Every Stage of the Chargeback Cycle,” this blog post provides a convenient summary recap.
The webinar addressed how eCommerce and mCommerce companies can win against chargebacks:
- Winning against chargebacks before an order is approved
- Winning against chargebacks after the transaction has been approved
- Winning against chargebacks with a frictionless payment experience
Don Bush kicked off his segment with two real-world examples that took place just 12 months apart to illustrate how rapidly fraud is changing (and how difficult it is for merchants, issuing banks, payment processors, and others in the e-commerce world to keep pace).
- In the first example from 2015, out of the 23 attempted fraudulent transactions, the issuing banks declined 19, and only four received credit card authorization. The techniques used by the fraudsters to disguise their identities were somewhat basic.
- The second example from just 12 months later exhibited a much higher level of sophistication. The techniques used to disguise the fraudsters identities were complex, varied, and more difficult to detect. As a result, out of 22 attempted fraudulent transactions, all 22 received credit card authorization. None were declined. In addition, the value of the attempted transactions was almost twice as high as in the example from the previous year.
Next, Don provided a brief overview of current fraud issues and their associated challenges. Don quoted Gartner VP, Avivah Litan about how these challenges required a multi-layered approach to fraud prevention:
Don emphasized that fraudsters look to exploit weaknesses in systems. When fraudsters get thwarted by multi-layer approaches, they give up quickly and move on to easier targets. Don detailed the essential capabilities necessary to a multi-layer approach, using the Kount platform as an example. The four main legs Don discussed:
- Real-Time Analytics: Multiple fraud screening technologies working in real time provide multiple hurdles, obstacles, and barriers for fraudsters. In addition, the integration of all these data elements adds an additional, cumulative layer: the “network effect” of Big Data.
- Powerfully Integrated Tools: The next layer in a multi-layer approach is provided via integrated access to best-in-class information providers, such as Ethoca, Experian, LexisNexis, TeleSign, Whitepages Pro, etc. This additional data provides critical context for even greater insight. Finally, integrating these multiple data feeds is key to ensuring that multiple and complex use cases are not a problem.
- Machine Learning: A third layer is artificial intelligence, which enhances the network effect of Big Data by using massive memory and computing resources to quickly recognize patterns otherwise undetectable to humans. Kount’s machine learning uses patented graph theory algorithms (PERSONA™) to extract highly predictive features in real time across the entire network. This provides deeper and more rapid identification of fraud as it emerges, unlike periodic off-line machine learning
- Skilled People as Your Advocates: A final layer is expertise in the fraud domain. It is key to informing and enhancing the other three layers in Kount’s multi-layer approach. Data scientist, fraud analysts, client success managers, and other experts – armed with best practices – provide the granular control needed to maximize sales, reduce fraud, and cut fraud mitigation costs.
Don closed with a brief summary of best practices to help merchants be successful at spotting fraud:
Best Practices Summary
- Audit your current system on an annual basis
- Most companies should outsource their fraud management
- Look into your own data to determine normal behavior
- Place multiple layers of technology in the way of fraudsters
- Employ a data-driven solution
- Consistent and ongoing training of staff
To wrap up the first segment, Melayna conducted a poll asking attendees: What is your top chargeback concern? The results of the poll:
36% - The costs associated with chargebacks
35% - Decreasing my chargeback rate
27% - The ratio of accepting good orders vs. bad
2% - Getting out of an excessive chargeback program
Keith Briscoe of Ethoca opened the second segment of the webinar with an alarming statistic about acceptance and false positives: according to Javelin, for every $1 dollar in fraud detected, merchants wrongly turn down $13 in good orders. The cause? Greater caution by merchants in response to increasing levels of fraud.
Keith then addressed five top chargeback challenges (see list below). He focused his remarks on how efforts to fight increased fraud can lead to poor customer experience, increased friction and lower acceptance rates.
- Direct Fraud Losses – Lost Goods & Services
- Rising Chargeback Processing Costs
- Poor Customer Experience & Increased Friction
- More Fraud Leads to Increasing Issuer, Decline Rates & Lower Acceptance
- Proliferation of Fraud Losses Through Repeat Attacks
A key challenge to addressing these five chargeback challenges is the information gap that exists between merchants and issuers. Issuers and merchants both possess strong fraud tools. The problem is that their data and insights exist in two separate worlds and the intelligence is typically not shared. The impact of this is that both needlessly suffer from higher fraud losses, lower acceptance and a poorer customer experience.
Ethoca’s solution? Create an out-of-band communication channel that transmits alerts about confirmed chargebacks to merchants in hours or even minutes. Here’s how it works: When a cardholder confirms fraud, an issuer in the Ethoca network – instead of processing a chargeback – notifies Ethoca, who then transmits an alert to the merchant in near real time. This rapid alert system provides a number of benefits to the merchant and the issuing bank.
Keith ended his segment with a summary of the challenges and problems solved by Ethoca and the Kount integration:
- Fight common fraud and chargeback challenges
- Single, integrated system for pre-authorization fraud scoring/screening and post-authorization chargeback mitigation/fraud recovery = higher overall acceptance
- Enhanced efficiency with automated matching – eliminate manual processes and transaction searches
- Quickly spot & stop fraud trends – integrated link analysis and VIP lists
- Put confirmed fraud and chargeback data to work faster – Datamart reporting
Mike Misasi of BlueSnap opened his segment with a series of provocative questions: “How would you run your business differently if fraud and chargebacks were not in your way? Would you sell in more countries? Would you accept more types of orders? Would you say ‘yes’ to riskier transactions?”
His point: fear of fraud prevents merchants from fully maximizing order acceptance and sales. This led to a discussion of abandonment and how to combat it in order to convert more sales. Mike explained that beyond the conventional model of shopping cart abandonment, there is the larger concept of “checkout abandonment” in which three factors negatively affect conversion rates:
- Friction – how difficult is it for customers to get to an order confirmation? Too many fields, the requirement to create an account, or too much time once the customer clicks the “Place Order” button all lead to higher friction and higher abandonment.
- Confidence – does the buyer feel safe? Websites that look illegitimate, or employ practices that make the transaction seem less than secure – for example, forwarding the customer to a 3rd party website -- lead to reduced confidence and reduced conversion rates.
- Payments – are there obstacles to authorizing the transaction? Orders can be declined by the issuer, by the merchant’s fraud system, or due to a technical error. Once again, conversion rates are negatively impacted.
Mike summarized best practices for minimizing friction, improving acceptance, and assuring payments in order to increase orders, increase sales and still fight fraud:
- Make sure your refund policy is clear and accessible
- Clearly communicate the order total and any recurring charges
- Use a billing descriptor that the shopper will recognize
- Make your fraud decision quickly
- Right size fraud prevention for your business