Skip to content

Instantly share code, notes, and snippets.

@bendalton

bendalton/fraud.md

Created March 18, 2019 15:32
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save bendalton/272bc17f9f641ddd3957754b1e3c4b21 to your computer and use it in GitHub Desktop.
Save bendalton/272bc17f9f641ddd3957754b1e3c4b21 to your computer and use it in GitHub Desktop.
Download ZIP
CATS notes
Raw
fraud.md

CATS - Fraud Prevention

Fundamentally automated fraud detection is hard. Disabling the immediate redemption capability is the best initial step which we've taken.

For ongoing fraud detection/prevention, providing an approval process with tools to manually identify potential fraud and take actions to restict that user's access.

What have we done

  • prevent backdating trips from before registration
  • Ban specific users ip addresses and have added new infrastructure to support banned users
  • Manually processing these trips
  • Begun work on disabling users who are suspected of fraudulent behavior

What are we going to do?

  1. Suspected Fraud Network - w/ popup and program (for astro) w/ recurring program email?
  • add excluded networks to incentives
  1. Prevent use of single device on multiple accounts (probably 2) (some nuance)
  • banned device ids
  1. Weekly/monthly reports with volume and change
  2. Point Program Approval Dashboard
  • yellow/orange highlights
  • gmail/hotmail/etc highlighted as questionable
  • ip addresses highlighted for multiple occurences and location
  • disable user popup - puts them in fraud network
  • (optionally) flag tx as potentially fraudulent popup (flag boolean, flag note for filtering)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment