> For the complete documentation index, see [llms.txt](https://docs.limephone.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.limephone.io/ai-receptionist/call-types-and-lead-qualification.md).

# Call types and lead qualification

Pick the right role for your receptionist and teach it what a promising caller looks like.

Not every business wants the same receptionist. A plumber wants messages taken; a sales team wants callers qualified. In LimePhone you steer this with the agent's **role** and its **instructions**.

## Choose a role

When you create an agent (**AI Receptionist → Create receptionist**), step 1 asks for a **Role**:

* **Virtual Receptionist** — greets callers, answers basic company questions, takes detailed messages.
* **Customer Support** — helps existing customers, troubleshoots, handles complaints with empathy.
* **Sales Assistant** — answers product/service questions and captures interested prospects.
* **Inbound Sales** — guides callers who are already interested toward a next step.
* **Lead Qualifier** — asks callers about their needs, budget, and timeline, and gathers details for your sales team.
* **Appointment Scheduler** — collects preferred dates, contact details, and the reason for the appointment.

Choosing a role auto-generates the agent's greeting and instructions around your business name and industry (from **Settings → Business info**). You can edit everything it generates — the role is a starting point, not a lock-in.

## Teach it what "qualified" means

There's no separate qualification settings screen — qualification lives in the agent's **instructions**. Edit them to spell out what you want asked and captured. For example:

```
Before ending the call, always collect:
1. The caller's full name and callback number
2. What they need (repair, quote, new installation)
3. Their timeline — this week, this month, or just researching
4. Their suburb or area

A caller is a strong lead if they need work within a month and are in
our service area. For strong leads, say a team member will call back
within the hour.
```

Be concrete: name the questions, the order, and what counts as a good fit for you. The agent asks naturally in conversation rather than reading a checklist.

## Where the answers end up

What the agent learns on a call lands in its **summary and transcript**. Review them in **Call Logs** (see [Call Logs & activity](https://docs.limephone.io/calls-and-messages/call-logs-and-activity)) and in the **Recent calls** feed on the AI Receptionist page, which shows each call's outcome — such as *Message taken* or *Handled* — at a glance. Caller details also flow into **Contacts** so you can follow up.

## Run different agents for different jobs

You can create **multiple agents** — for example a Lead Qualifier on your sales number and a Customer Support agent on your main line — and connect each to different numbers in step 4 of the wizard.

Verify your questions actually get asked with a test call: [Test your AI Receptionist](https://docs.limephone.io/ai-receptionist/test-your-receptionist).