Fraud score
A fraud score is a numeric or qualitative risk rating for a given IP or visit, built from several signals at once — IP reputation, proxy/VPN/datacenter markers, and anomalies in device fingerprint or behavior. The higher the score, the less likely the visit is a safe, genuine human rather than a bot or a checker.
What feeds into the score
IP-intel providers and anti-fraud systems roll datacenter and proxy/VPN lists, mismatches between IP geo and the claimed location, and behavioral markers — actions that happen too fast, no mouse movement, unusual click patterns — into a single number. The system then uses that score to decide whether to let the visit through to the offer or route it to a safe page.
Using it in a live funnel
The usual setup is a threshold: cross it, and the visit gets sent to a white page instead of the offer automatically, protecting against moderators, checkers, and wasted spend on traffic that was never going to convert anyway. Sensitive verticals often set a tighter threshold and combine fraud score with geo and device filters. APEX's cloaking bakes IP-intel risk scoring directly into its filter rules, with no separate third-party service to wire up.
FAQ
- Does a high fraud score mean it's definitely a bot?
- No — it's a probabilistic risk estimate, not a hard fact, and it's meant to drive cloaking rules rather than serve as a final verdict.
- Do you need to manually review high-fraud-score visits?
- Usually not — cloaking rules route that traffic to a safe page automatically, and manual review is only useful when investigating a specific incident.
Cloaking, anti-bot, push, split tests and your own domains — in one service.
Get started free