How this works in Reputoo
First-response timing on every thread
The clock starts at the customer's first inbound message and stops at the first human reply. Average first response is reported per agent alongside total conversations and resolved count, so a slow queue is visible before customers complain about it.
SLA breach counter
Conversations that exceed the threshold are counted as breaches on the team performance view rather than quietly ageing at the bottom of a list.
Escalation out of the bot
When the AI agent cannot resolve a thread, or the customer asks for a person, the conversation escalates to the human queue with its full history. An unanswered escalation is the thing SLA tracking is there to catch.
Off-hours handling
Outside business hours the AI agent answers rather than leaving the thread cold, and anything it cannot close is queued for the morning. This is usually a bigger win than a faster alert, because the customer gets an answer instead of a notification going to a sleeping manager.
