Guide
Choosing between AI, hybrid and human answering
Updated
The choice is usually presented as a technology decision and it is closer to a pricing decision about your own calls, which is a much easier question to answer honestly.
Talk to a specialist What each model handles, what happens at the edge, who is accountable, and what it may do outbound.
What is a missed call worth?
Take the value of an average job or order, multiply by the proportion of enquiries that convert, and you have the cost of a call that goes unanswered.
For a trade business or professional service that number is frequently large. That is what makes any answering arrangement pay for itself, and the reason to stop optimising the per-minute rate.
What proportion of calls are routine?
Listen to fifty calls, or read fifty transcripts, and sort them: routine, needs judgement, would upset somebody if handled badly.
That split decides the model far better than any vendor demonstration. Mostly routine points at automation with escalation. A large judgement tail points at people.
Will anybody own the queue?
The in-house-with-an-assistant model is the cheapest on paper and the most likely to be abandoned, because it needs somebody internally to watch the captures and act on them.
If you cannot name that person, choose an outsourced model instead. A capture nobody reads is a missed call with a longer transcript.
Then ask the four unglamorous questions
Show me the escalation trigger list. What hours does the human layer actually cover, as distinct from the AI layer? How do I review transcripts and find the calls that should have escalated and did not? And can this system initiate outbound calls, is that off by default, and can I verify that it is off?
Those four separate a serious vendor from a demonstration. None of them is about the voice.