AI that helps you convert more leads—without losing your personal touch.
AI-powered Conversation Agents inside Alsona allow you to automate intelligent replies, handle objections, nurture leads, and book meetings on your behalf—directly from your LinkedIn inbox. This guide walks you through how to create your agent, what each field means, and how to use it strategically for outreach success.
Where to Create Your AI Conversation Agent
You can create an agent in two different places:
1. Inside a Campaign
When creating a new campaign, you’ll find the option to Set up an AI Conversation Agent in the campaign settings.
Simply select “Create New Profile.”
2. From the Bottom-Left Menu
If you want to build your agent first (before assigning it to a campaign):
Go to the bottom-left menu inside your Alsona dashboard
Select AI Conversation Agents
Click “Create New Profile” in the top-right corner
Both paths bring you to the same setup page.
Step 1:
Step 2:
Step-by-Step Guide: Creating Your AI Conversation Agent
Below is a breakdown of each field you’ll complete—along with tips to get the best results.
1. Basic Information
AI Profile Name
Choose something descriptive, such as:
Outbound – SaaS Demo Booking
Recruiting – Software Engineers EU
Agency – B2B Lead Gen Discovery Call
This helps you stay organized if you use multiple profiles.
2. Core Message
This is a one-sentence summary of what you're offering. This should be your elevator pitch.
Examples:
“We help B2B teams generate qualified demos using AI-powered outreach.”
“I’m connecting because we’re hiring senior developers and your background stood out.”
“We help agencies scale outreach with multi-seat automation.”
3. Challenges
Explain the pain points your prospect faces. Feel free to be descriptive so that the AI really understands the challenges.
Examples:
“Sales reps spend hours sending manual follow-ups.”
“Hiring quality candidates is slow and competitive.”
“Agencies struggle to deliver predictable lead flow.”
4. Approach
Explain how your offer solves these challenges. Again, be descriptive so that AI has a lot to pull from when creating a response.
Examples:
“Our AI handles outreach and replies so reps can focus on closing.”
“We connect engineers with vetted remote-first companies in under 10 days.”
“Our automation platform handles engagement, nurturing, and workflows for you.”
5. Additional Instructions
This is one of the most powerful fields—because it lets you teach the AI exactly how to behave.
You can add brand voice, tone, restrictions, cultural details, industry nuance, compliance needs, and more.
Examples of great Additional Instructions:
A. Brand Voice & Style
“Sound friendly, concise, and confident. Avoid corporate jargon.”
“Keep replies warm but professional. Use short paragraphs.”
“Use a consultative tone—never pushy or salesy.”
B. Local Customs / Region-Specific Behavior
“For prospects in Germany, maintain formal tone and avoid first names unless they use mine first.”
“For Japanese prospects, keep messages highly polite and avoid overly casual statements.”
“For UK prospects, avoid American idioms and keep tone understated, not overly enthusiastic.”
C. Relationship-Based Selling
“If the prospect seems hesitant, share a soft-value question such as: ‘Happy to share a quick example if helpful?’”
D. Compliance Preferences
“Never promise results or guarantees.”
“Avoid discussing pricing unless the prospect asks.”
E. Industry-Specific Instructions
“For tech leads, reference integrations and APIs when relevant.”
“For HR prospects, emphasize time savings and candidate quality.”
F. Unique Creative Uses
Persona Switching: “Sound like a senior strategist, not a junior rep.”
Humor Control: “Add light, subtle humor only when appropriate.”
Confidence Levels: “Keep tone assertive but never aggressive.”
Scarcity Cues: “If they ask about availability, subtly mention limited onboarding slots.”
Qualification Rules: “If the prospect is outside the US/EU, always ask what region they sell to before proposing next steps.”
You can build incredibly sophisticated agent behavior with this field.
6. Goals & Fall-Back Goals
Choose what you want the AI to achieve:
Examples of Primary Goals
Book a meeting
Gather qualification details
Share more product information
Build rapport
Reconnect with inactive leads
Examples of Fall-Back Goals
If they say “not interested,” shift to learning why
If they need more info, offer a short explanation
If they decline a meeting, send a soft follow-up in 24–48 hours
You can also add custom instructions explaining how the agent should pursue each goal.
7. Settings & Automation
Message Length
Choose from:
Standard (75–120 words) – Most natural for outreach
Tone of Voice
Options include:
Professional
Friendly
Direct
Conversational
(Choose what matches your brand)
Respond to automatic emails?
If enabled, the AI will reply to:
Out-of-office responses
Auto-confirmations
Calendar invites
Handle objections or negative replies?
If enabled, the AI will send thoughtful follow-up messages like:
“Totally understand — what tools are you currently using today?”
“No problem. Mind if I share a quick example relevant to your industry?”
Receive meeting link
“Ask them to share their meeting link first”
Useful if you want prospects to pick the time.
8. Generate a Sample Conversation
Before saving, Alsona lets you generate a sample conversation. This will show you exactly how the AI would respond to a real prospect.
You can adjust tone, instructions, or fields until you’re happy with it.
9. Saving & Assigning Your Agent
Once you're satisfied:
Click Save
You can assign the agent to a campaign immediately
Or assign it later from your AI Agents list
How the Agent Works in Your Inbox
Once assigned:
✓ It replies on your behalf inside your LinkedIn inbox
Replies are labeled “Auto-sent” so you always know which messages came from the AI.
✓ You can manually take over at any time
If the AI hasn’t replied yet, you’ll see an option to “Take Over the Conversation”.
✓ Response Window
The AI replies within 1–4 hours of the prospect’s message to maintain authenticity and natural timing.
Final Tips for a High-Performing AI Agent
Use specific, not generic instructions
Add cultural or regional nuance if targeting multiple countries
Include objection-handling guidance
Keep your core message simple and focused
Revisit your sample conversations often and refine
Create separate agents for different industries or personas
Don’t forget to test different tones (Professional vs. Conversational)