How does ClearSale help businesses prevent false declines?

ClearSale’s strategy to prevent false declines starts with understanding which transactions are typically “good” and “bad” for each client.

Companies have to juggle providing exceptional customer experience (CX), preventing fraud and optimizing revenue. It can be a difficult undertaking, and too often, companies find out that they’ve turned away good customers long after any opportunity to win them back.

Run an auditing program

When we engage with clients, one of the first things we do is run an auditing program to measure the accuracy of the company’s existing automated rules. We do this by taking a control group of the client’s previously processed orders and comparing that group with a random set of transactions to determine if the right choice was made. The audit validates any existing rules and pinpoints the adjustments needed to improve accuracy.

Our client success team also works with clients to find out what data is missing to help determine the types of transactions that should be approved and declined.

Utilize AI and machine learning

We utilize an AI-enabled algorithm that leverages trends, intelligence and data gathered from decades of fighting fraud in the most high-risk regions of the world. The client-specific data is also used to “teach” our system which of their transactions should truly be considered fraudulent.

Using this technology, we can automatically approve or decline most orders quickly with a high level of accuracy. Instead of declining suspicious orders, we can flag them for contextual reviews.

Expert contextual reviews

The small percentage of flagged orders (about 2%-3%) are evaluated by our more than 2,000 fraud analysts who have the experience to recognize some of the hardest-to-spot fraud patterns. If necessary, our analysts may reach out to customers, but they do so in a way that is in line with exceptional CX.

Leverage new data

The additional data gleaned from those contextual review is leveraged to help our system distinguish valid transactions from fraud with even more accuracy. This continues over time, with our system becoming “smarter” as we process more and more of the client’s transactions – which increases their approval rate and revenue.