Containment Rate: definition and how to read it
Containment rate is the share of calls an automated system finishes without handing them to a human. It measures how much the AI handled, but not whether callers were actually satisfied with the outcome.
Also known as: self-service rate, automation rate, resolution without escalation
How it is calculated
Divide calls ended by the automated system, with no transfer to a person, by total calls received in the period. Decide up front how to count callers who hang up mid-way: a caller who gives up looks 'contained' but was not served. Many teams exclude calls abandoned before the agent responded, and track abandonment separately.
Why it can mislead
High containment can mean the AI is good, or that it is hard to reach a person. The healthier pairing is containment together with resolution: did the caller's issue get fixed, as shown by a post-call check, repeat-call rates or an outcome tag? Containment on its own rewards agents that refuse to escalate.
Example
A support line contains 70% of calls, but 20% of those callers ring back within a day. The effective containment is lower than reported. Linking calls by caller number over 24 to 48 hours reveals this.
Where this shows up at Vaami
Vaami's voice intelligence outputs call outcomes and transcripts that teams can use to evaluate how many calls the agent completes and how many need a person.
See it on a real call.
Hear how a Vaami voice agent handles this on a live conversation in your language and your use case.