Signup fraud prevention
One call at account creation, before the account exists to defend.
What the problem actually is
Every downstream fraud problem is cheaper at registration. Once an account exists it accrues content, referrals, sessions and support history, and removing it becomes a decision with collateral damage. Screening at creation is the only point where rejection costs nothing.
Signals that carry the weight here
IP, email, phone and device weighted together. Any single signal in isolation produces far more false positives than the combination.
A confirmed-abusive value is never averaged away by clean signals elsewhere — that averaging is how single-signal tools miss obvious fraud.
Where to start
strictness=1 (balanced) — Balanced. Route review verdicts to email or phone verification rather than rejecting them.
A defensible starting point, not a guarantee. Strictness moves the verdict thresholds only; the risk score itself never changes, so you can re-tune without re-scoring anything.
The call
curl -X POST "https://www.layercall.com/v1/score/user" \
-H "X-Api-Key: YOUR_KEY" \
-H "Content-Type: application/json" \
-d '{"ip":"$IP","email":"$EMAIL","phone":"$PHONE","device_id":"$DEVICE_ID"}'1,000 lookups a month free, no card. Get an API key.
Questions
How much latency does this add to signup?
Typically under 300ms, and it is one call rather than four. If it matters to your flow, call it asynchronously after account creation and act on the verdict rather than blocking the form — the signal is just as useful a second later.