Symptom or Need

It is identified that the majority (or 100%) of records in an outbound dialing campaign fail to establish effective contact, showing the following indicators:

• Dominant Call Statuses: Busy, Failed, or Congestion.
• Penetration: Very low or zero (0%).
• Behavior: High volume of dialing attempts without generating effective contacts.

Context / Scenarios

This incident usually occurs due to operational changes or infrastructure limitations, such as:

• Activation of new campaigns or changes in the SIP provider/trunk.
• Use of aggressive dialing modes (Aggressive Predictive or Turbo Dial) or increased dialing speed.
• Server overload (shared server or multiple campaigns on the same Skill).
• Low-quality databases (poorly formatted numbers, blank records, or unvalidated prefixes).
• Network instability or latency toward the telephony provider.

Key Concepts to Consider: 

• No Answer: The customer does not pick up.
• Answer Machine: An automated machine answered the call.
• Congestion / Busy: May occur due to:

           - Issues within the customer’s carrier network.

           - Poor signal reception on the customer’s end.

           - Massive outages from the customer’s carrier.

• Abandon: The customer answers but hangs up before being transferred to an agent (due to signal issues or customer choice).

Response / Solution

1. Initial Validations and Statistics

      Gather basic data to scope the failure:

• Identify the operation name, percentage of impact, and number of affected agents.
• Data Analysis: Calculate actual penetration, review average attempts per record, and determine the "Dominant Result" (Is it mostly Busy, Failed, or Congestion?).
• Compare behavior across different providers.

     2. Controlled Testing

• Perform 5 to 10 manual calls to the numbers that failed in the campaign.
• Validate if calls go through correctly using an alternative provider (if available).
• Confirm which provider is currently routing the calls.
• Verify that the Caller ID is authorized by the provider.

      3. Infrastructure and Provider Review

• Server: Monitor CPU usage (>80%) and RAM at the time of the event. Validate if Dials Per Minute (DPM) exceed the server’s capacity.
• Connectivity: Validate latency and connection stability with the SIP provider.
• Carrier: If the issue is "Congestion," try switching providers to rule out spam blocking or carrier saturation.
• Shared Server: Confirm if the server is shared with other operations and evaluate the performance impact.
• Simultaneous Campaigns: Review the number of active campaigns and their resource consumption.
• SIP Configuration: If applicable, consider changing the Siptar server on the outbound trunk to optimize call stability.

      4. Parameterization Adjustments

• Reduce predictive dialing aggressiveness or disable Turbo Dial (if applicable).
• Adjust dialing speed based on the actual number of active agents.
• Database Cleanup: Validate prefixes, remove blank records, and ensure that already managed numbers are not being reprocessed.
• Confirm that multiple campaigns are not associated with the same Skill to avoid call assignment overhead.
• Evaluate switching providers (e.g., TIGO, CLARO, etc.) to rule out service blocking or degradation.
• Validate if the issue is isolated to a specific carrier.
• Confirm potential restrictions due to traffic volume or dialing flagged as spam.
• Verify number portability status in Colombia in case of routing failures.

Possible Causes 

The Busy / Failed / Congestion statuses can stem from various factors. The primary causes are detailed below according to the affected component:

A. Infrastructure (Server)
This is related to capacity limitations or environment overload:

• Sustained CPU usage exceeding 80%.
• High spikes in Dials Per Minute (DPM).
• Insufficient RAM.
• Simultaneous execution of multiple campaigns on the same server.
• High latency toward the SIP provider.
• Shared server with high resource consumption.

Typical Result: Predominance of Congestion statuses.

B. Provider / SIP Route

Associated with external or routing failures:

• Carrier saturation.
• Caller ID restrictions.
• Issues with the configured trunk.
• Impact on a specific operator.
• Portability issues (Colombia).
• Incorrect route or unauthorized provider.

Typical Result:

• Mostly Congestion → Possible provider saturation.
• Mostly Failed → Route or authorization issues.

C. Campaign Parameterization

Related to internal configurations or database quality:

• Dialing speed exceeding actual capacity.
• Overly aggressive predictive dialing.
• Turbo Dial activated without sufficient resource capacity.
• Insufficient number of active agents.
• Multiple campaigns using the same Skill.
• Incorrect prefixes.
• Blank or poorly formatted records.
• Reprocessing of already managed numbers.

Typical Result:

• Mostly Failed → Configuration or database issues.
• Mostly Busy → Database saturation or excessive retries.

D. Agent Connectivity

Related to agent availability and stability:

• Agents not truly available.
• Internal network issues.
• High TCP Delay.
• Occupancy level higher than estimated.

Typical Result:

• Call assignment failures.
• Dropped calls during handling.

Recommendations

For proper incident management, it is recommended to:

• If Congestion predominates: Prioritize reviewing the provider and server capacity.
• If Failed predominates: Validate configuration, SIP route, and numbering format.
• If Busy predominates: Check for database saturation and the number of retries.
• Do not escalate to the provider without first conducting manual tests.
• Do not increase dialing speed without previously validating CPU, RAM, and DPM (Dials Per Minute).
• Compare behavior across providers if more than one is available.
• Always document the exact time of the event to facilitate technical analysis.
• Maintain evidence of tests performed before escalation.