Using Our Custom Bot AI Agent to Generate a Description
🤖 What the AI Bot Is For
The AudienceLab AI bot is your shortcut to building clean, high-performing custom model descriptions — fast.
It helps you:
- Get high-intent topic suggestions
- Write one-paragraph descriptions using our Clarity5 framework
- Control lens and granularity
- Remove confusion, brand stacking, or persona noise
Whether you're starting from scratch or refining a known idea, the AI bot ensures your models are sharp, structured, and submission-ready.
🧭 How to Use the AI Bot to Build a Custom Model
✅ Step 1: Ask the Bot for a Suggestion
If you’re not sure how to frame your idea, start by asking the bot for a suggestion.
What to say:
“Give me a suggestion for people looking to refinance their home. Use the Solution lens. Granularity 3.”
The bot will return 2–4 specific, lens-based topic ideas.
Do:
- Keep it simple — describe the intent or need
- Specify a lens (Brand, Product, Function, Service, Solution, or Event)
- Set granularity from 1–5 (5 = most specific)
Pro tip: The bot will ask:
“Would you like this to be more specific or less specific?”
Use this prompt to fine-tune the idea before writing the paragraph.
✅ Step 2: Choose Your Favorite Suggested Topic
Pick the topic that best matches the exact behavior you want to model.
Example interaction:
You say:
“Give me a suggestion for cold email. Lens: Function. Level 4.”
Bot suggests:
- Inbox warm-up automation
- Domain rotation for sender reputation
- Engagement protection strategies
You choose:
“Domain rotation for sender reputation”
Now you're ready to generate the model description.
✅ Step 3: Ask the Bot to Write the Model Description
Once your topic is chosen, ask the bot to generate a clean paragraph using Generation Mode.
What to say:
“Write a custom model for: domain rotation for sender reputation. Use the Function lens. Granularity 5.”
The AI bot will return a one-paragraph description using the Clarity5 framework:
- Problem
- What it does
- Where/how it’s used
- Real-world signals
- No fluff, no comparisons, no personas
Example output:
“Automated domain rotation systems that protect sender reputation, maintain cold email deliverability, and reduce blacklist risk for outreach teams scaling across multiple inboxes.”
This is now ready to submit.
✅ Step 4: Submit the Model
- Copy the paragraph from the AI
- Go to AudienceLab → Custom Models → Create New
- Paste it into the ‘Description’ field
- Click Submit
Note: AudienceLab scans over 200 billion URLs per week — but your model only works if your input is sharp. That one paragraph is the signal.
✅ Step 5: Add Filters After the Model is Built
Once your model is approved, you can apply filters to refine who you're targeting.
Use platform filters like:
- 🌍 Location (e.g., Florida)
- 🧑💼 Job title or department
- 🏭 Industry (e.g., Healthcare, SaaS)
- 💰 Revenue or employee size
- 🌐 Domain list (e.g., Dream 100)
Important: Never include these filters in the model paragraph itself. The model is about what someone is doing — the filters decide who they are.
🧪 Bonus: Ask for a Review or Rewrite
If you want feedback, ask:
“Can you review this for clarity and remove any noise?”
The bot will check for:
- Mixed brands
- Persona language
- Vague phrases
- Keyword stuffing or ambiguity
✅ Summary Workflow
- Ask the bot for a suggestion (intent + lens + granularity)
- Choose the most relevant topic
- Ask the bot to generate a one-paragraph description
- Paste into the AudienceLab Custom Model Builder
- Apply ICP filters (job title, geo, domain) after model creation
🔧 Example Prompts
To get a suggestion:
“Give me a suggestion for weight loss behavior. Function lens. Granularity 3.”
To generate a paragraph:
“Write a model for zero-down HVAC financing. Service lens. Level 5.”
To refine or review:
“Review this model and remove any noise or broad terms.”
Last updated on August 6, 2021