Book a Call
Skip to main content
Case study · Voice AI

AI Voice Bots for Insurance Outbound Calling, With Dialer Integration and Analytics

Built two outbound AI voice bots for insurance calling, connected them to the client's dialer through their own SIP trunk, and added a web dashboard to monitor them.

Phase 2 brief followed
Client
US insurance business
Industry
Insurance marketing
Timeline
Nov 2025 onward
Engagement
Fixed price, 2 milestones
How it works

The system, step by step

In this case study, we cover a voice automation project for an insurance client. The goal was two outbound AI voice bots, one for Medicare and one for final expense insurance, working inside the client's existing dialer.

  • VICIdial dialer

    Client's own campaigns

  • Dedicated SIP trunk

    Calls on their infrastructure

  • AI voice bots

    Medicare and final expense

  • Live agent

    Transfer when needed

The challenge

What the brief demanded

The client's brief set out requirements across voice, telephony and reporting:

The client

The client was a US insurance business running outbound calling campaigns on VICIdial. They supplied the call scripts and wanted natural-sounding AI conversations, routed through their own infrastructure, with clear reporting.

  • Natural-sounding voice interaction, using technology such as GPT-4 and ElevenLabs.
  • Medicare conversations covering information, plan comparison and enrolment guidance.
  • Final expense conversations covering policy options and discussion.
  • VICIdial integration, with a dedicated SIP trunk so outbound calls route through the client's own trunk.
  • A web dashboard tracking do-not-call numbers, dead calls, bot performance, live calls and transfers from the bot to a live agent.
What we built

Built and handed over

We built the two voice bots to the client's scripts, integrated them with VICIdial through a dedicated SIP trunk, and delivered the monitoring portal described in the brief.

  • A Medicare outbound voice bot and a final expense outbound voice bot.
  • VICIdial integration with a dedicated SIP trunk for outbound calling.
  • A web analytics dashboard for do-not-call tracking, dead calls, bot performance, live calls and AI transfers.

The project had a six-week target and ran in two paid milestones, with all milestones completed. A second-phase brief from the client followed in early 2026.

  • Large language model and text-to-speech technology (GPT-4, ElevenLabs or similar)
  • VICIdial
  • SIP trunking
  • Web analytics dashboard
More case studies

Other systems we have built

Sales automationAn AI Offer Engine and Outreach Loop on Airtable, GPT and ApolloProposal automationTurning Form Submissions into PPC Proposals with Jotform, ChatGPT and ZapierData pipelineAutomating PDF Retrieval and Data Extraction into a CRM for Lead Generation

Want a system like this?

Tell us the process that costs your team the most time. On a 30-minute call we will say what we would build and what it would take.

Book a Call