Deflection rate measures the percentage of customer inquiries handled without reaching a human agent, while resolution rate measures the percentage of inquiries that were actually resolved. The difference is avoidance versus outcome.
A contact counts as deflected the moment it stays away from an agent, even if the customer abandoned the chat or never got a real answer. A contact counts as resolved only when the issue is genuinely fixed. That gap is why a system can post a high deflection rate and a much lower resolution rate at the same time. Deflection tells you how much work the team avoided. Resolution tells you how much of the customer's problem got solved.
Deflection rate vs resolution rate at a glance
| Dimension | Deflection rate | Resolution rate |
|---|---|---|
| What it measures | Share of inquiries kept from a human | Share of inquiries genuinely solved |
| When a contact counts | On avoiding an agent, even abandoned | Only when the issue is fixed |
| What it rewards | Closing or avoiding contacts | Solving the right things well |
| What it tells you | Work avoided | Problems solved |
This distinction is central to how Aide, the agentic AI platform for customer experience, defines success. Stated soberly, deflection rewards a system for closing or avoiding contacts, while resolution rewards it for solving the right things well. Aide does not lead with deflection-flexing. It leads with resolution rate, and backs it with Intent Coverage Rate as the diagnostic underneath: the share of real customer demand, mapped from actual conversations, that has deployed and verified automation. That is resolution checked against the full picture of what customers ask for.
Automation must earn the word resolved. An intent goes live only after its behavior has been verified on real past conversations, so the resolution figure reflects problems fixed, not contacts dodged. And because that figure is measured against everything customers actually ask, automated or not, it cannot quietly exclude the hard parts. The question worth asking is not how much was deflected, but how much was resolved.
Frequently asked questions
- Can deflection rate and resolution rate be different?
- Yes, and they often are. Deflection counts every contact kept away from an agent, including abandoned ones, while resolution counts only contacts that were actually solved, so deflection can run well above resolution.
- Which metric should I optimize for?
- Resolution, not deflection. Deflection can be inflated by customers giving up, while resolution reflects whether the issue was genuinely fixed. Aide backs resolution rate with Intent Coverage Rate, a diagnostic that checks automation against real customer demand.