Hi.
Can Shopify’s fraud analysis detect duplicate accounts created to exploit rewards or referral programs? Are there apps that might be more effective in this regard?
Topic summary
Main issue: whether Shopify’s native Fraud Analysis can detect duplicate customer accounts created to exploit rewards/referral programs, and if there are dedicated “fraud apps” better suited for this.
What was shared: a link to Shopify’s Fraud Analysis documentation and suggestions of third‑party apps related to loyalty/referrals:
- Rise.ai: checks IP addresses and verifies first purchase to validate referrals.
- EasyAccounts: reportedly helps prevent referral abuse (link provided).
- Smile.io and Stamped.io: manage customer accounts and use unique referral links.
Clarification from the requester: they are not seeking loyalty apps with some fraud controls, but rather details on Shopify’s own fraud analysis or standalone fraud-detection apps. They note the documentation doesn’t explain how Shopify’s algorithm works or whether it flags duplicate accounts based on similar email, address, phone, or IP.
Status: No concrete answer yet on Shopify Fraud Analysis’s capability to detect duplicate accounts; the core question remains unresolved and the discussion appears open.
Hello JaviExpress,
Shopify Fraud Analysis documentation
Shopify Help Center | Fraud analysis
Apps for Fraud Prevention:
Rise.ai
This app specializes in gift cards and loyalty store credit. It ensures that referrals are genuine by checking IP addresses and verifying that it’s the referrer’s first purchase at your store.
EasyAccounts
It can help prevent referral abuse
Smile.io and Stamped.io
These loyalty and rewards apps allow you to manage customer accounts and create unique referral links
Thanks!
But I’m not interested as much in apps that include fraud control as I am in ‘fraud apps’ or Shopify’s fraud analysis itself.
I don’t think the documentation in Shopify explains how the fraud algorithm works or if duplicated accounts (e.g. two buyers have similar email, address, phone, ip, etc.) would be flagged as fraudulent