Skip to content
Start a project
Our Work  /  Case study
Voice AI · GoHighLevel · GPT-4o

Voice AI Receptionist with Demo-on-Demand Follow-Up

An inbound AI receptionist that answers questions from a knowledge base, books discovery calls, writes caller data to the CRM and triggers an outbound call when a demo form is submitted.

The brief

A service business needs every enquiry answered immediately, whether it arrives by phone or through a form, and needs the conversation to end with clean CRM data and a booked next step. The agent has to sound like a calm, helpful team member, stay inside the services it can talk about, and handle callers who are sceptical because they get a lot of spam.

What we built

  • A voice agent with a defined role, tone and behaviour rules: casual and calm, does not repeat the caller’s name, stays on topic and answers personal questions naturally before returning to the goal.
  • A knowledge base attached to the agent with a written rule for when to use it, so business answers come from approved information.
  • Phone and availability settings: a dedicated number, the AI agent as backup to the number, and optional working hours.
  • Eight in-call actions: two workflow triggers, one appointment-booking action against a discovery-call calendar offering three days of dates, and five contact-field updates (email, running ads, ad budget, follow-up and first name).
  • An outbound workflow: when the “Demo On Demand” form is submitted, a Voice AI outbound call starts automatically.

How it works

  1. Visitor submits the demo form, or calls the business number.
  2. The agent greets, qualifies and answers questions using the knowledge base.
  3. Answers are written to contact fields as the call happens.
  4. The agent offers calendar slots and books the discovery call.
  5. Follow-up workflows continue the conversation after the call.

Results

We publish measured outcomes only when the client has approved them. Delivered scope and configuration are shown below; ask us for verified results on a similar project.

Inside the build

Screens from the delivered system

Select any image to enlarge it.

Agent details: name, voice, language, timezone and LLM model
Agent details: name, voice, language, timezone and LLM model
Agent goals: knowledge-base rule and role and behaviour prompt
Agent goals: knowledge-base rule and role and behaviour prompt
In-call actions: workflows, booking and contact-field updates
In-call actions: workflows, booking and contact-field updates
Phone numbers, backup mode and working hours
Phone numbers, backup mode and working hours
Form-triggered outbound call workflow
Form-triggered outbound call workflow
Start a project

Have a build in mind? Let’s scope it properly.

Tell us what you are building, which platforms are involved and where it is currently stuck. You get a clear technical plan before any work starts.