This feature introduces comprehensive event logging in Application Insights that captures exactly how average wait time is computed, evaluated, and shown to customers.
End-to-end average wait-time diagnostic events
The system logs detailed AWT diagnostic events in Application Insights at two key moments:
When a conversation begins and the system evaluates whether average wait time should be shown.
When the availability APIs return the average wait-time value (if applicable).
Each event includes standard metadata (channel type, organization ID, conversation ID, scenario, and subscenario) and feature-specific fields such as calculated average wait time in seconds, whether average wait time was shared with customers, and reason. This allows administrators to pinpoint the exact cause when average wait time is missing or is an unexpected value.
Historical analytics for average wait-time accuracy
Supervisors can compare:
Average wait-time values shown to customers.
Actual wait times experienced by conversations.
This helps teams validate prediction quality, identify discrepancies, and tune overflow thresholds.
Business Value:
Average wait time (AWT) is one of the most visible and impactful signals in a contact center, used to manage customer expectations and trigger overflow logic. But when average wait time doesn't appear or seems incorrect, supervisors often lack the visibility needed to understand why.
Average wait-time diagnostics brings transparency and control to average wait-time calculation by giving administrators and supervisors detailed diagnostic logs, historical analytics, and real-time queue level average wait-time values that they can use to troubleshoot issues and validate system behavior without needing engineering or support.
The feature dramatically reduces explainability gaps, increases trust in the system, and helps contact centers make data-driven staffing and overflow decisions.
Enabled for:
Enabled for admins/makers/marketers automatically
GA date: August CY2026