Fraud
How Invitebase decides fraud automatically, the evidence behind each decision, and the block lists you manage — with no manual review queue
The Fraud surface is where you read Invitebase's automatic fraud decisions and manage the block lists they draw on. Deterministic rules decide on the pass that validates a referral, and the outcome is final. There's no manual review queue: referrals resolve on their own, so this surface is about visibility, not dispositioning each one by hand.
How referrals resolve
A referral that clears the rules approves and moves toward fulfillment, after any configured hold window. One that trips a rule is rejected — never paid, never billed. There is no third outcome and nothing to wait for: the decision lands the moment the referral validates. See Fraud prevention for the controls behind it.
The evidence on each referral
Every decision shows what the system knew when it decided, so a rejection is never a black box:
- Why it resolved as it did — the rule that fired, as a reason code, plus the block-list signals behind it.
- Referrer standing — tenure, past validated referrals, and prior rejections.
- Referral velocity — this referrer's referrals over the last day and week, and referrals across the whole app sharing the same click identifier.
- The validation record — the immutable gate evaluations that validated the referral, and the rewards it produced with the status of each.
Block lists
Block lists are the one control you edit here: emails, device identifiers, and referrer accounts. Blocked identities stop being attributed at ingestion, and a block-list hit on a referral that reaches validation rejects it outright.
Analytics
The fraud chart shows the consequences: the approved-versus-rejected mix over time, and the total reward spend your controls blocked before it was paid.
Test-mode referrals flow through the same rules and the same decision so you can rehearse safely, but they never bill and never move money. See Environments and test mode.