What is Deflection Rate?
The percentage of support questions resolved by self-service — typically an AI assistant — without creating a ticket or reaching a human agent.
If 1,000 customers ask questions in a month and 620 get their issue resolved by the assistant without filing a ticket, the deflection rate is 62%. Multiplied by cost per ticket, deflection is the most direct way to express a support AI's financial impact.
The honest version of the metric matters. Counting every AI interaction as a deflection inflates the number; counting only confidently-answered questions that did not subsequently become tickets reflects reality. Deflection should also never come at the cost of trapping users who need a human — a low-confidence question that escalates quickly is a success, not a failure.
Deflection rate pairs naturally with gap analysis: the questions that fail to deflect are exactly the list of content to write next.
How BeforeQuery uses this
BeforeQuery meters resolved-versus-escalated outcomes per workspace, applies confidence scoring so only well-grounded answers deflect (in chat and the Support Form Agent), and lets you replay historical tickets through simulations to estimate deflection before going live.
Related terms
Hallucination
When a language model generates fluent, confident content that is factually wrong or entirely invented.
Citations
Links from an AI answer back to the specific source documents used to generate it, allowing users to verify every claim.
Knowledge Base
A curated, indexed collection of an organization's knowledge — docs, wikis, tickets, repos — that an AI assistant retrieves from when answering questions.
From definition to deployment
Connect a knowledge source and see cited AI answers on your own content in minutes — free, no credit card required.
Get Started Free