Retail Fraud

Point Predictive has developed a highly predictive analytic model that helps retailers accurately identify those high-risk applicants who are most likely to default due to application fraud or EPD, while impacting the smallest numbers of good customers at the point of sale. Our Data Scientists use machine learning and artificial intelligence techniques in combination with retail data to minimize application fraud and EPD while preserving a great customer experience at the point of purchase.

There are many advantages for retailers to offer credit at the point of sale for large retail purchases, such as appliances. Because the consumer is in your store or on your website ready to make a large purchase, it is more important than ever that the credit experience be as positive as possible. It is critical that retailers be able to make accurate credit decisions quickly, without taking on excessive risk from application fraud or customers who use their real identity, but never have any intention to repay the loan, typically referred to as early payment default (EPD).

Retail Fraud Manager is able to detect more than 75% of risky applications in the highest scoring 10% of applications. This means retailers can dramatically reduce fraud and EPD without disrupting 90% of customers with an additional fraud review.

Retail Fraud Manager employs predictive analytics and adaptive machine learning where self-learning models adjust to new patterns of fraud quickly. We leverage a retailer’s internal data in combination with consortium data to build the best possible retail online fraud transaction fraud model.

Retail Fraud Manager models can be deployed on order processing systems or hosted by Point Predictive. Data Scientists and Business Consultants help you deploy the best model to reduce false positives and increase retail online transaction fraud detection.

With the US adoption of CHIP cards, fraudsters have evolved their focus for stealing from retailers. They are using ever-more sophisticated Card Not Present (CNP) and other online transaction schemes, as well as using people known as “mules” to pick up items in stores that have been purchased fraudulently online. Retailers, as a result, need more powerful ways to detect fraud and keep merchant charge-backs from the credit card associations in check. Our approach has been found to reduce manual reviews and automatic rejects by 30% or more. It also lowers your cost of data.