# Building a Sales AI Agent

### 🧭 Overview

A **Sales AI Agent** in raia is designed to qualify leads, handle prospect questions, and automatically route qualified opportunities to human sales reps. It works across **SMS, Email, and Voice**, and can be deployed through:

* **Outbound Campaigns** using **Mission Control**
* **Trigger-based communication** via your CRM or other apps
* **Inbound handling** of texts, emails, or phone calls

Once set up, the agent runs autonomously—handling early-stage outreach, qualifying leads, logging interactions, and keeping your sales team informed in real time.

***

### 🧱 Phase 1 – Agent Setup in Launch Pad

#### 1. Create a Sales Agent

In **Launch Pad**, click **Create Agent** and:

* Select the **Sales Assistant** role
* Define:
  * Agent Name (e.g. “AI SDR – North America”)
  * Description (e.g. “Qualifies inbound and outbound leads across all channels”)
  * Tags (e.g. Sales, SDR, Outbound, NA Region)

***

#### 2. Write Clear Instructions

Instructions define the agent’s goals, tone, and logic.\
Keep them simple — like prompting ChatGPT.

**Example Instructions:**

> You are a Sales AI Agent for ACME Corp.\
> Your job is to qualify leads, answer product questions, and pass good leads to a human sales rep.\
> Ask 2–3 qualifying questions.\
> If the lead is qualified, notify the sales rep.\
> If unsure, log the conversation and ask for help.

***

#### 3. Upload Sales Content

Train your agent with:

* Product datasheets
* Pricing and plan comparisons
* Case studies and testimonials
* Common objections and rebuttals
* Qualification checklists
* Demo booking workflows

Use **raia Connect** to:

* Upload PDFs, Docs, Slides
* Convert them into clean, structured training data
* Push directly into the AI’s knowledge base

Organize by use case or product line for best results.

***

### 📡 Phase 2 – Enable Multi-Channel Communication

In the **Skills** tab, enable:

#### ✅ SMS Skill

* Assign a local or toll-free Twilio number
* Set welcome/fallback/opt-out messages
* Ready for outbound and inbound communication

#### ✅ Email Skill

* Use `@raiabot.com` or connect your domain
* Set from-name, signature, autoresponder
* AI can send and respond like a sales rep

#### ✅ Voice Skill

* Assign a phone number (via raia or Twilio)
* Create greeting (text-to-speech or audio file)
* Enable transcription and optional IVR

***

### 🧠 Phase 3 – Qualify, Score, Route

#### 🔍 Use the Scoring Skill

* Auto-score every conversation based on:
  * Positive signals
  * Prospect intent
  * Qualification answers
* Define scoring criteria in setup (e.g. "Interested + Budget + Timeline = Qualified")

#### 🔔 Use the Notification Skill

* Instantly alert a human rep when:
  * A conversation begins
  * A conversation ends with a high score
* Alert via **email or SMS**

#### 🌐 Use the Webhook Skill

* Push all conversation data into your:
  * CRM (HubSpot, Salesforce, Zoho, etc.)
  * Google Sheets
  * Marketing automation tools
* Trigger on `OnThreadEnd` or `OnMessageEnd`

***

### 🚀 Phase 4 – Deployment Options

#### Option 1: **Outbound Campaigns**

Use **Mission Control** to:

1. Upload a list of leads
2. Choose SMS, Email, or Voice
3. Write a message template
4. Launch the campaign
5. AI handles responses, qualifies leads, scores conversations

#### Option 2: **Trigger from CRM or App**

* Set up webhook or API triggers from:
  * CRM field updates
  * Form submissions
  * Marketing engagements
* Agent is automatically invoked to start a conversation

***

### 📥 Phase 5 – Handle Inbound Engagements

Your Sales AI Agent can automatically respond to:

* **Inbound SMS** → Instant reply with qualification flow
* **Inbound Emails** → Responds like a human rep
* **Inbound Phone Calls** → Answers, asks qualifying questions, routes if needed

***

### 🧪 Phase 6 – Test and Train in Copilot

* Simulate conversations in **Copilot**
* Add feedback (👍/👎, edit, retrain)
* Watch for low-score threads to improve performance
* Use **Takeover Mode** to let reps jump into a live convo with a hot lead

***

### ✅ Final Setup Checklist

| Task                                             | Status |
| ------------------------------------------------ | ------ |
| Agent created in Launch Pad                      | ✅      |
| Instructions defined clearly                     | ✅      |
| Sales documents uploaded & tagged in **Connect** | ✅      |
| SMS, Email, Voice Skills enabled                 | ✅      |
| Webhook Skill pushing to CRM                     | ✅      |
| Notification Skill alerting human reps           | ✅      |
| Scoring Skill configured with thresholds         | ✅      |
| Outbound or CRM-based triggers tested            | ✅      |
| Copilot tested + human takeover enabled          | ✅      |

***

### 🔁 Example Workflow

1. **CRM adds a new inbound lead**
2. Webhook triggers the Sales AI Agent
3. Agent reaches out via SMS or Email
4. Prospect replies → AI qualifies them
5. Conversation scores 9/10 →
   * Sales rep notified via email
   * Transcript pushed to CRM via webhook
6. Rep follows up with high context

***

 


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.raiaai.com/dev-program/agents/building-a-sales-ai-agent.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.