Field Notes
The enquiries your care business never hears about
A practical way to measure out-of-hours demand before deciding whether a voice-assistant prototype is worth exploring.
The Sunday evening problem
Families don't decide about care between nine and five. The conversation may happen at the kitchen table on a Sunday evening, and somebody may pick up the phone there and then. If that call goes to voicemail, some people will not leave a message. Because the missed need is hard to see, the first job is to measure it rather than assume it.
What a prototype would need to prove
A useful voice-assistant prototype would need to capture a caller's purpose clearly, explain its limits, collect only the details needed for a callback, and hand anything sensitive or uncertain to a person. It should be tested with representative scenarios before any operational trial. Summaries would need routine human review, with consent, retention, and escalation rules agreed in advance.
This is not about replacing staff
The narrow idea is to explore calls that a team may otherwise miss, such as evenings, bank holidays, or times when everyone is helping someone. The human work would still belong to humans: listening, calling back, understanding the situation, and deciding what happens next. Low call volume may also mean that a clear voicemail greeting remains the simpler choice.
Test the theory before spending anything
You can check whether the problem exists without buying a system. Ring your own business after 6pm and on a Saturday morning, and listen to what a prospective caller hears. Then count missed calls and voicemails for a fortnight. If the number is material, define a small validation plan. If it is tiny, the tally has still provided a useful answer.