Detect interactions where the agent actively attempts to retain or save a customer who has expressed churn intent, such as offering incentives, discounts, plan changes, or addressing concerns to prevent cancellation.
This evaluation considers both agent attempted customer retention and service cancellation intent expressed event to build a complete picture of what happened in the conversation.
Customer intent is often expressed indirectly, buried in the flow of a conversation that is ostensibly about something else. A billing inquiry becomes a churn signal. A support call reveals an upsell opportunity. These moments are invisible in traditional QA because scorecards are not designed to look for them.
Detecting retention save attempt across your entire interaction volume transforms what used to be anecdotal observations into measurable intelligence. Operations leaders can see trends, segment by customer type or agent team, and connect conversation signals to business outcomes.
Compass analyzes the full context of the conversation to determine whether retention save attempt occurred. This is not keyword matching or phrase detection. The evaluation considers meaning, sequence, and conversational dynamics to distinguish genuine instances from surface-level similarities.
The evaluation is calibrated to account for ambiguity. When the evidence is not strong enough to make a confident determination, the signal surfaces as unclear rather than being forced into a binary present-or-absent result. This means teams can trust the signal when it does fire.
Operations leaders use this signal to understand what customers are actually telling them, at scale. Individual interactions become data points in a larger picture of customer needs, friction, and intent.
Retention and CX teams use it to trigger proactive workflows. When the signal fires, it can inform follow-up actions, routing decisions, or escalation paths, turning a reactive service model into a responsive one.
Product and strategy teams use aggregated signal data to identify systemic issues. If this signal spikes for a particular product, segment, or time period, it usually means something changed upstream that needs attention.
This signal is part of Chordia’s Signal Intelligence capabilities.
