I’m trying to figure out why the SIP trunk failure metrics in the Performance Dashboard do not align with the raw conversation data. We have a Genesys Cloud CX instance (EU-West) integrated with a NICE CXone SIP trunk for outbound campaigns. The Architect flow is straightforward: originate call, wait for answer, transfer to agent. No complex scripting or IVR logic prior to the SIP request.
The issue arises when comparing the ‘SIP Trunk Status’ widget against the ‘Agent Performance’ view. The dashboard reports a 2% failure rate for the trunk, citing ‘Remote Party Busy’ as the primary cause. However, when exporting the conversation details for the same 24-hour period, only 0.5% of calls show a ‘Busy’ disposition. The remaining 1.5% are simply missing from the successful call logs, yet they do not appear in the error logs either. They seem to vanish between the SIP invite and the final disposition record.
This discrepancy impacts our billing reconciliation with the carrier. We need to know if the Performance Dashboard aggregates SIP-level rejections differently than the conversation outcome engine. Is there a latency window where SIP failures are counted in the dashboard but not yet finalized in the conversation detail view?
How should we reconcile SIP-level failure counts with conversation-level dispositions for accurate billing?