Sangoma CX Technical Guide - Mitigating Spam Tagging on Legitimate Calls

Legitimate calls being marked as Spam is one of the most difficult problems that the telecom industry is facing right now. The industry reached this situation because of the rise in fraud calls and the lack of compliance to the FCC regulations from some companies in relation to outbound calling and more specifically the use of automated dialers and robocalling.

 Based on these circumstances, telco operators took action to protect their clients by implementing their own analytics engines to identify what it is a spam versus a legit / compliant call. Sadly these analytics engines can be very aggressive when it comes to marking DIDs as Spam Likely, and the tag can be applied after making only a few outbound phone calls. This can cause any number of issues for legitimate businesses that are simply trying to make outbound calls to their clients.

Contact centers, due to its operation, are areas heavily affected. The purpose of this guide is to explain how this issue occurs, and what can be done by organizations to mitigate the problem.

Metrics that can Potentially Tag a Number as Spam

While the metrics that are monitored can vary from carrier to carrier, these appear to be the common variables weighed against DIDs and what Sangoma recommended to minimize the risk of being tagged as spam.

  • Quantity of outbound phone calls:

    • Outbound call volumes that exceed 150 calls per day on a single DID/Caller ID may convince inbound carriers that you are a robocaller and may trigger an alarm.

    • Recommendations:

      • Do not exceed more than 150 calls per day per Caller ID. If the organization needs to do a high volume of calls, use a pool of owned DIDs/Caller IDs and cycle them.

      • Do not consistently use the same set of DIDsCaller IDs daily for high volume outbound calls. Let them rest several days giving time for the algorithms to reset the counters. A proper cycling management of the Caller ID pool is key.

  • Quantity of calls sent to Voicemail:

    • This is considered by the carrier as if the contact called did not identify who is calling. A common behavior when someone is receiving a call from an unknown source is forcing the call to voicemail or waiting until the call hangs or goes to voicemail.

  • CPS (Calls Per Second):

    • Making many simultaneous calls might trigger the engine to consider those calls are being done by an automated dialer if a high CPS is consistent during the day.

    • Keep in mind if the CPS is very large even in a short time period it maps a robodialer profile and these systems can only be used if the organization complies with specific requirements and this will also trigger an alarm.

  • Average Length of Call:

    • It is a very good metric to confirm the calls generated are really ending in real conversions. If the Average Call Length is low and the number of calls dialed is high, it might trigger a Spam alarm.

  • Quantity of calls 6 seconds or less:

    • This is considered by the engines as a call abandoned by the caller, and it is a technique commonly used by organizations using predictive dialers. These systems dial more calls than available agents so it is frequent to have contacts that pick up a call that cannot be handled by an agent and the most common resolution is dropping the call. As with robodialers, these systems can only be used if the organization complies with specific requirements

  • Making calls Before/After Business Hours:

    • Federal and State regulations define when outbound calls can be made. These regulations could differ in some States. This can be easily reviewed by the engines and will trigger an alarm.

 All these metrics share a common ground with existing regulations, so it is extremely important to comply with FCC regulations such as TCPA. Also keep in mind these algorithms could also be comparing the numbers contacted with the national DNC (Do Not Call) list. Consider besides having a Caller ID blocked, not following FCC regulations might result in expensive settlements and/or fines.

 It typically takes 60-90 days to remove the Spam flags in numbers. This does not guarantee once removed being tagged again a few days later if the Analytics engines considered it again as Spam. It is very important to keep track of the metrics monitored, have a pool of Caller IDs and implement a Caller ID cycling strategy to minimize the risk of being blocked.

The recommendation is to also frequently review the contactability metric on all the DIDs used for outbound calls. This metric will help to identify if a DID might get or has already been tagged as Spam. Detecting if the number got tagged requires historical data to compare. If the contactability dropped and nothing changed, it is a good indicator the number got blocked by one or many telcos.

Supplemental Actions

Remediation for spam likely tags on B2C conversational traffic is typically resolved by using resources such as www.FreeCallerRegistry.com, however, contact centers may not see positive results with these methods. These actions should be considered supplementary steps to the suggestions that were listed in the section above.

 This free portal streamlines the caller registration process with First Orion, Hiya, TNS and registers the calling business with the Analytics Engines that support major wireless and wireline carriers in the US.

When should organizations register with the Free Caller Registry

Sangoma recommends that customers register their DIDs as soon as they are purchased. If you are porting numbers to your network, we recommend that you add registration with this database to your porting workflow after submitting the order to Sangoma.

What to do if a number has been tagged as Spam Likely

If your numbers have already been tagged as Spam, we recommend opening a help desk ticket with the carrier that has applied it. Help Desk links for T-Mobile, Verizon, AT&T, and several other websites are included below.

Other ways to mitigate Spam Likely tags

There are a few other steps that you can take to mitigate Spam tags:

  • Ask customers to save your business number:

    • Some third-party scam blocking apps allow users to block calls from unknown numbers that aren’t listed in their contacts, so your call may not reach them at all if one of those services are in use.

    • If customers are saving your number in their contacts, this will reduce the chance of that happening, while also making them more likely to ignore any spam likely tags since they will recognize your number.

  • Ask clients to adjust or remove third-party spam blockers:

    • While this solution may seem impractical, it could alert your customers to potential problems they have with other blocked calls. Some users often have spam blockers installed unknowingly.

Return to Documentation Home I Return to Sangoma Support