Phone System
Review and Proposal . Confidential

Inbound calls: what is happening, and what to do

A full review of the inbound phone system, and an interim rebuild we can do inside Vonage without waiting for the Twilio migration.

Call data: 1 to 13 August 2026, thirteen consecutive complete days, 8,117 inbound voice calls and 2,016 call recordings. . Page updated: 14 August 2026

47.8%
of callers reach a person
56%
of voicemails get no callback in 72 hours
76%
of call volume is people calling back
41.8%
of our outbound calls show clients the main menu number

The short version. Our call volume is not a staffing problem. A twelve-second recording tells every existing client to press 0, which routes about 297 people a day through two members of staff, whose onward transfers ring for five seconds and then drop into mailboxes that are deleted after being emailed. Clients call back, which is why three quarters of our volume is repeat callers.

Reception is not the problem. They answer a higher share of the calls that reach them than any other group in the firm. They have nowhere reliable to send anyone.

Start with the proposal. The current-state map is the evidence underneath it, for anyone who wants to check the working.

Start here

Proposal
Three options, one front line, no dead ends
What we propose changing and in what order: the new menu, one front-line team that verifies identity and warm transfers at most once, every destination becoming a visible queue rather than a group, and direct lines failing over to their team instead of to voicemail. Includes the build sheet, how we will measure it, and what it does not fix.

The evidence

Current state
Where 8,117 calls actually go
The full map. Every entry point and what it answers, the calls that reach nobody, what callers are actually ringing about, callback rates across both phone systems, the numbers nobody dials, and how the Newlin and Fleschner traffic is entangled.

A note on how to read these. Everything is counted from call records unless it is explicitly labelled as a judgment. Two figures are judgments and are marked as such: how much a front-line team could resolve without transferring, and which calls a menu could have routed. Both come from reading call recordings, not from measurement, and both are worth challenging.

Powered by Novact.ai