Fraud Types
Spreading illegal call volume thinly across a large pool of rotating calling numbers so no single number triggers a blocking threshold.
The tactic borrows its name from snowshoe spam: distribute weight across a wide surface so you never break through. A campaign may use thousands of DIDs, each placing only a few dozen calls before being retired.
Per-number reputation systems are structurally weak against this, because each number is discarded before it accumulates a reputation.
Defence requires looking at the customer and the behaviour pattern across the whole number pool, not at individual numbers — which is exactly what per-number blocklists cannot do.
An automatically dialed call that delivers a pre-recorded or artificial voice message, legal only with the right consent or exemption.
Analytics-driven tags such as 'Scam Likely' or 'Spam Risk' that terminating carriers and handsets apply to inbound calls.
The rate at which new call attempts are offered to a network, and one of the clearest early indicators of an automated dialing campaign.
The percentage of call attempts that are actually answered, used as both a route-quality measure and a fraud indicator.
SipShield scores live SIP traffic with patented AI so the controls behind this term run automatically — and leave the audit trail regulators expect.