Submit Deal

referDot Pro

Fraud review

Flag suspicious activity, track client risk scores, and resolve issues before paying rewards.

Pro docsSetup guide

Overview

Fraud review is a simple queue for your team to document and track suspicious referral behavior. It complements Rewards — use it before you approve or mark conversions as paid.

  • Console: /#your-slug/fraud
  • Access: Owner, Admin, and Staff (fraud:read)
  • Dashboard shows Open fraud flags (unresolved items only)
  • Manual workflow — referDot does not auto-block payouts based on flags today

Using fraud review

  1. Open Fraud in the workspace sidebar.
  2. Pick the client when the issue is about a person. The menu shows their score, risk label (Clear, Watch, Review, Hold), and how many open flags they already have.
  3. Enter a reason and severity from 1 to 10. Severity is added to that client's score.
  4. Add a note on the flag card. Notes stay on the client and show on later flags too.
  5. Investigate in Clients, Code lookup, or Rewards as needed.
  6. Click Resolve when handled, or Reopen if the issue returns.
  7. Share with support files the flag, client, score, and notes as a new support ticket. The ticket link stays on the flag so you do not send it twice.
Example · Duplicate redemption

Staff notice code 5550100 redeemed twice under different names → Admin adds flag “Two names, one phone” severity 3 → Rewards manager keeps related conversions PENDING → Resolve after callback.

Fraud flags

Each flag is a review ticket with:

  • Reason — short note (max 200 characters)
  • Severity — 1–10 scale for prioritization
  • Status — open or resolved
  • Client link — optional; when set via API, shows name, code, and fraud score

The console lists the latest 100 flags. Resolved flags stay in history for audit; only open flags count toward the dashboard metric.

Client fraud score

Every client has a fraud score (starts at 0). When you create a flag with a clientId through the API, that client's score increases by the flag severity (1–10 per flag).

Score bands: Clear (0), Watch (1–3), Review (4–7), Hold (8+). The page shows the band and open flags next to the client. Scores do not automatically reject lookups or conversions.

Recommended workflow

With rewards

Keep conversions PENDING while related fraud flags are open. After review, Approve or Reject, then Mark paid only for approved items. See Rewards documentation.

Weekly hygiene

Check dashboard Open fraud flags → resolve or escalate → scan high fraudScore clients before batch payouts.

Example · Self-referral check

Lookup shows new customer registered under advocate’s code twice in one week → flag severity 5 via API with clientId → hold conversions → reject duplicate if confirmed.

API reference

Authenticated workspace APIs (same session as the console):

  • GET /api/pro/orgs/{slug}/fraud — list flags
  • POST /api/pro/orgs/{slug}/fraud — create { reason, severity?, clientId? }
  • PATCH /api/pro/orgs/{slug}/fraud — { id, resolved }

Full payloads and admin score override: docs/pro-fraud.md in the repository.

What it does not do

  • No automatic fraud detection on lookup or conversion create
  • No webhook events when flags are added or resolved
  • No auto-hold on payouts when score is high — your process decides
  • No email or SMS alerts for new flags

For conversion automation (not fraud), see Pro webhooks.

Troubleshooting

  • Can't access Fraud — your role may lack fraud:read (unusual for default Owner/Admin/Staff).
  • Flag has no client details — created from the UI without a client link; use API with clientId next time.
  • Score didn't change — score only increments when clientId is sent on create.
  • Dashboard count won't drop — resolve open flags; resolved items don't count.
  • Need to reset a score — contact referDot support or use admin tools at /admin/pro/[orgId].