Tagging an incoming call with a warning such as "Spam Likely" or "Scam Risk" based on analytics, so the recipient can decide whether to answer.
Terminating carriers and device or app vendors use analytics engines to score incoming calls. Inputs include STIR/SHAKEN verification results, attestation level, call volume and duration patterns, consumer complaints, and the reputation of the calling number and signing provider. Calls above a threshold are labeled, sent to voicemail, or blocked outright.
Labels are probabilistic, and legitimate callers are sometimes caught. High-volume outbound programs such as appointment reminders, collections, and political or survey calls are especially vulnerable if their numbers are new, rotate frequently, or draw complaints. Even a modest share of complaints against a number can be enough to tip it into a warning label on some networks.
Different terminating networks use different analytics, so the same number can be labeled on one carrier and display normally on another. Businesses that believe they are mislabeled can typically request a review through each carrier or analytics provider. Because labels are reviewed separately on each network, resolving a mislabeling problem can take time and persistence, and prevention is far easier than cleanup.
Why it matters to carriers
For originating providers, call labeling is where your customers feel the consequences of authentication and reputation. Ensuring legitimate customers receive accurate attestation, helping them register numbers with analytics providers, and watching for sudden drops in answer rates can head off complaints and protect enterprise accounts from churn.