Deflection rate
Deflection rate is the percentage of customer issues resolved without human agent involvement. The most-quoted metric in AI-for-support sales.
Deflection rate is the percentage form of ticket deflection: (tickets closed without human) / (total incoming tickets). It is the metric every AI-for-support vendor leads with and the metric most likely to mislead a buyer if used in isolation.
The arithmetic is simple; the interpretation is not. A 60% deflection rate can mean “60% of customers got the right answer fast” or “60% of customers gave up and went to Twitter.” The dashboard treats them identically.
Published benchmarks: Zendesk enterprise median 41.2%, top quartile 58.7%, with a 30-40 percentage-point gap between vendor marketing claims and field reality. Intercom Fin: ~41% average, 65% peak. ServiceNow Virtual Agent: 45-60% range cited, with 80-94% in best-case deployments.
The honest reading: deflection rate is a useful diagnostic when paired with AHT, FRT, CSAT, and recoverability. As a standalone contracted metric, it incentivizes the wrong behavior — close the ticket without resolving the issue.
Why deflection rate matters in 2026
The 2025-2026 wave of AI in customer service has shifted the conversation around deflection rate from feature checklist to operating outcome. Vendor research consistently documents a gap between marketing claims and field reality — Zendesk's CX Trends 2026 puts the gap at 30-40 percentage points across the category — and that gap shows up wherever deflection rate is part of the deployment conversation.
For support teams evaluating vendors today, the question is rarely whether the vendor offers deflection rate; it's whether the vendor will contract on the outcomes deflection rate is supposed to produce. Outcome-contracted models (deflection, AHT, FRT, CSAT in the SOW) shift the risk profile compared to feature-access models (per-seat or per-resolution pricing). The choice between the two is often the most important architectural decision in the program.
Read more in the POV essay Native helpdesk AI is built for safe defaults for the structural argument on why deflection rate alone is not enough to move outcomes, and Deflection is the wrong goal — outcomes are for what to ask for in the contract instead.
Frequently asked questions
Roughly yes — tickets closed without human / total tickets — but the ‘without human’ threshold varies. Some vendors count any auto-acknowledgement as deflection.
Auralis publishes ~60% deflection in repetitive categories as the steady-state cohort baseline. The contract includes the outcome metrics that determine whether that deflection moves the business or carries hidden cost.
Put AI to work for your support team
See how Auralis deploys custom AI agents in days, not months.