---
Title: "Which Metric Measures Tickets Avoided by AI Self-Service?"
Url: "https://computer.io/answers/which-metric-measures-tickets-deflected-by-ai"
Topic: "Metrics"
Description: "The metric is ticket deflection rate: the share of requests resolved by AI or self-service without becoming a ticket for an agent. Formula and caveats."
---

# Which metric measures the percentage of tickets avoided due to AI or self-service?

The ticket deflection rate measures the percentage of support requests avoided through AI or self-service. It is calculated as the number of requests resolved without a human agent divided by the total number of requests, times 100. Pair it with AI resolution rate and reopen rate so you count solved problems, not customers who gave up.

## Formula

Deflection rate = (requests resolved by AI or self-service ÷ total support requests) × 100

## Related metrics

| Metric | What it adds |
|---|---|
| Self-service score | Help center views ÷ tickets created |
| AI resolution rate | Share of AI conversations that solved the issue |
| Containment rate | Share of bot conversations not handed off (can include abandons) |
| Reopen rate | Catches false resolutions |
| CSAT on AI | Quality check |

## FAQ

### What is a ticket deflection rate?

The percentage of support requests resolved without a human agent, through AI or self-service.

### Is containment rate the same as deflection rate?

Similar, but containment usually measures bot conversations that weren't handed off, which can include customers who gave up.

### What is a good deflection rate?

It depends on your product and channels. Track trend and pair it with CSAT and reopen rate.

## Related terms

- [Deflection rate](https://computer.io/glossary/deflection-rate)
- [Ticket deflection](https://computer.io/glossary/ticket-deflection)
- [AI resolution rate](https://computer.io/glossary/ai-resolution-rate)
- [What AI resolution rate should you expect for L1 tickets?](https://computer.io/answers/what-ai-resolution-rate-to-expect-for-l1-tickets)

## How Computer helps

- [Customer agent](https://devrev.ai/solutions/customer-agent)

## Further reading

- [Ticket deflection](https://devrev.ai/blog/ticket-deflection)
- [Metrics for knowledge base](https://devrev.ai/blog/metrics-for-knowledge-base)