For reporting/analysis of actual response times of support agents when working on Cases it would be useful if the system would track time even after SLA breach/failure.

As an example, when an agent provides a first response to a customer before failure time of the SLA instance, system is calculating and registering the actual duration between creation and success of the SLA instance.

When the SLA breaches, system is registering the total time between creation and failure time which is obviously always the value as specified in the SLA item.

When a support agent, after SLA breach does provide the first response to the customer, system does not act on that.

But from reporting perspective, it is important to understand what the actual response times of agents have been, even if the SLA failed. This will provide better insights in the overall/average responses to support cases.