Multi-Segment Service Level Metrics with Adjustable Targets
A
Adeel
Customer Context (Binance)
CRM: Intercom
Customers using bot handoffs in Intercom need to measure service level performance at different stages of the conversation lifecycle to:
- Distinguish between queue wait times vs. agent response times
- Identify whether bottlenecks are due to understaffing (queue wait) or agent behavior (response delay)
- Separate accountability for capacity planning vs. individual agent performance
Currently, Assembled only supports one service level metric per channel/queue, which combines all stages into a single measurement.
Requested Functionality
Three Independent Service Level Types:
- Chat Wait Time Service Level
Formula: cs_join_time - bot_trans_time (Intercom fields)
Meaning: Time customer waits in queue after bot transfers to human agent queue
Use Case: Measures staffing adequacy
Target: Configurable per queue (e.g., 80% under 5 minutes)
Status: "achieved" if waiting_time < target, else "breached"