An unresolved issue is what happens when a conversation ends but the customer’s problem doesn’t. The customer called or contacted support with a specific need — maybe their service isn’t working, maybe they need account information, maybe they have a billing question — and despite time spent on the interaction, that need remains unmet when the conversation closes.
This signal identifies interactions where the customer’s original concern was left hanging. Not calls that were difficult or complex, but specifically those where the resolution status is incomplete. The customer may have been told to try something that didn’t work, or given partial information, or had their issue acknowledged but not addressed.
The signal uses context suppression to focus on genuine resolution failures rather than interactions where the customer chose not to proceed with a resolution that was offered.
Unresolved issues don’t disappear when the call ends. They generate repeat contacts, escalation calls, and complaints. A customer whose billing question goes unanswered will call back — often frustrated that they have to re-explain everything. An unresolved technical issue becomes a service cancellation.
The operational cost is immediate: every unresolved issue typically becomes at least one additional contact. But the relationship cost compounds over time. Customers remember when their problems weren’t solved, and those memories influence every future interaction.
Traditional QA often misses this because scorecards focus on process compliance, not outcomes. An agent can follow every procedure perfectly and still leave the customer’s issue unresolved. This signal bridges that gap by tracking what actually matters: whether the customer’s problem got solved.
Compass evaluates whether the customer’s stated concern was addressed by the end of the interaction. This goes beyond checking whether an agent provided information or followed troubleshooting steps — it assesses whether the underlying issue that brought the customer to contact support was actually resolved.
The signal accounts for context to avoid false positives. If a customer chooses not to proceed with a repair appointment that was offered, or decides not to accept a resolution that was provided, the interaction isn’t flagged as unresolved.
Operations leaders track resolution rates as a leading indicator of repeat contact volume. Teams with high unresolved rates consistently see higher call volumes and longer handle times, as customers circle back with the same issues.
Supervisors use unresolved flags to identify training needs. An agent who consistently leaves issues unresolved may need coaching on problem-solving techniques or access to additional resources to handle complex cases.
QA teams prioritize unresolved interactions for review. These are the calls where process improvement matters most — understanding what prevented resolution helps refine procedures and agent training.
This signal is part of Chordia’s Signal Intelligence capabilities.
We'll walk you through real interactions and show how each signal traces back to specific conversational evidence — so your team can act on what actually happened.
