Content Analytics and Event Analytics are two different approaches to analyzing and detecting spam...
STIR/SHAKEN Solves the Spoofing Problem, but How Do You Stop the Bad Behavior?
When we say the Secure Telephony Identity Revisited (STIR) and Secure Handling of Asserted information using toKENS (SHAKEN) standard, or STIR/SHAKEN, solves the spoofing problem, that statement comes with its caveats. STIR/SHAKEN validates whether the source of an incoming call is authorized to use that number. The attestation data it provides (full, partial, or gateway) serves simply as a clue about caller intent. That clue still needs to be interpreted by artificial intelligence (AI) to make a reliable determination about the caller’s intent. In other words, just because your home security includes a camera doesn’t mean all break-ins will be thwarted. The real value of that security system is in its analytical ability to tell the difference between a stray cat walking by the front door and an unfamiliar human stealing a package off the porch.
In this article, we cover:
- How STIR/SHAKEN contributes to robocall mitigation.
- The value of content-based analytics.
- Using brand reputation management to stop bad behavior.
Did you know that tech-support scams alone cost an average of $450 per individual? Unfortunately, STIR/SHAKEN isn’t an on/off switch for preventing these and other types of voice scams and spam. That means not all calls signed with A-level attestation will automatically be delivered. Similarly, not all calls signed with B-level attestation will be blocked. Carriers and providers must employ call analytics tools, such as those offered by YouMail Protective Services, to gain the call blocking and spam identification features they need to complete the unwanted-robocall-blocking loop.
Good thing too, because robocalls are back on the rise. After an initial dip in robocall rates immediately following the STIR/SHAKEN deadline a year ago, the new year has seen a return to form. In our latest data, we saw an 8.5 percent increase in robocalls between May and June 2022.
How Call Analytics Leads To Robocall Mitigation
In the realm of conversational AI, there are event-based analytics and content-based analytics. The event-based variety uses call data to predict outcomes. An algorithm uses data such as answer seizures and call duration rates to predict fraud or spam. These predications, however, can only be proven post-fact.
Content-based analytics examines the actual call content to determine what occurred. So even though event-based methods surface trends and clues, they naturally lead to more false-negatives. Content-based analytics, on the other hand, almost never result in false-positives since they focus on call payload data, or the actual communication delivered in a call.
Call analytics technology, such as that offered by YouMail Protective Services, uses algorithms processing event-based data and content-based data to score every call for voice service providers and enterprises. Scoring determines call treatment, such as whether to block, label, or redirect calls.
How To Stop Bad Behavior
If STIR/SHAKEN isn’t a silver-bullet solution, is there one at all? The short answer is no. Because the most effective robocall blocking technology takes attestation, and event- and content-based analytics all into consideration when scoring calls, the onus falls on enterprises to engage in identity and brand reputation management to supplement their robocall mitigation strategy. This includes detection and elimination of imposter traffic to protect brand image, company reputation, and customer relationships. Enterprises must seek zero-hour identification of unwanted robocalls to weed out bad actor campaigns that sully their otherwise positive brand reputation.
Brand reputation management should include web scanning tools that detect misuse of logos, domain names, and other assets. It should also monitor postings on social media, and phishing and vishing (voice phishing) attempts that use brand names in attacks over email and phone lines. By taking action against brand impersonation attacks, an enterprise will ensure that its good behavior is rewarded with high call analytics scoring on legitimate call outreach programs, and bad behavior is punished with low scoring on imposter campaigns.
Because that is the ultimate goal, isn’t it — to deliver the best possible consumer experience by successfully engaging in legitimate brand marketing and support outreach without fear that they will instead choose to ignore your call because they’ve been burned so many times by fraudsters before. And while STIR/SHAKEN has added an invaluable tool for voice service providers and enterprises to leverage, it is limited in mitigating unwanted robocalls on its own.
To reduce risk from robocalls, service providers and enterprises must learn the critical role of audio analytics — such as content-based analytics — in robocall threat mitigation programs. Download our free white paper on STIR/SHAKEN and call analytics today.