Case Study: StackCommerce achieves 25% drop in chargebacks and 5x ROI with Sift

How StackCommerce reduced chargebacks and saved valuable time and money

StackCommerce, a leading native commerce platform powering deal stores for publishers, was losing time and money to credit-card fraud on instantly delivered digital goods. Their legacy rules-based fraud system didn’t scale as order volume grew, forcing large manual review queues, mass approvals that increased disputes, and mounting chargebacks.

They implemented Sift’s machine-learning fraud platform with automated Formulas and Actions, integrating in under two weeks and training the model to improve accuracy. The result: a 25% drop in chargeback rate (about $2,000/month saved), orders up 30% with no additional hires, manual review reduced to one person spending ~2 hours/day, and nearly 5x ROI.

StackCommerce

Brandon Robbins

Product Manager