Overview

AI agents are the heart of Marketrix. They’re intelligent assistants that can help your users through three powerful interaction modes: Show, Tell, and Do. This guide will help you understand what agents are and how they can transform your customer support.

​

What Are AI Agents?

An AI agent is an intelligent assistant powered by advanced artificial intelligence that can understand natural language, access your knowledge base, and provide helpful responses to your users.

​

How Agents Work

​

Agent Capabilities

Your AI agents can:

• Understand natural language - They can interpret questions in plain English
• Access your knowledge - They use your documents and videos to provide accurate answers
• Provide personalized help - They can adapt their responses to your business context
• Handle multiple interactions - They can maintain context throughout conversations
• Learn and improve - They get better over time as they interact with more users

​

Three Interaction Modes

Marketrix agents can help users in three different ways:

​

Tell Mode - Answering Questions

What it does: Provides information and answers questions Examples:

• “What are your business hours?”
• “How do I reset my password?”
• “What features does your product have?”
• “Can you explain how this works?”

Best for:

• Customer support questions
• Product information
• Policy explanations
• General inquiries

​

Show Mode - Demonstrating Tasks

What it does: Shows users how to perform tasks with step-by-step guidance Examples:

• “Show me how to create an account”
• “Show me how to use this feature”
• “Show me how to navigate your website”
• “Show me how to set up the product”

Best for:

• Tutorials and guides
• Feature demonstrations
• Process explanations
• How-to instructions

​

Do Mode - Performing Actions

What it does: Actually performs tasks on behalf of users Examples:

• “Help me fill out this form”
• “Guide me through the checkout process”
• “Automate this task for me”
• “Complete this process for me”

Best for:

• Form filling
• Account setup
• Process automation
• Task completion

​

Agent Types

​

AI Agents

What they are: Intelligent AI-powered assistants Characteristics:

• Powered by advanced AI technology
• Can understand and respond to natural language
• Use your knowledge base to provide accurate answers
• Can handle complex conversations and follow-up questions
• Available 24/7 without breaks

Best for:

• Customer support
• Sales assistance
• User onboarding
• General information

​

Human Agents

What they are: Placeholders for human support staff Characteristics:

• Represent real human support staff
• Can take over from AI agents when needed
• Provide human expertise and empathy
• Handle complex issues beyond AI capabilities
• Available during business hours

Best for:

• Escalation from AI agents
• Complex technical issues
• Sensitive customer situations
• High-value customer interactions

​

How Agents Interact with Users

​

Natural Language Understanding

Your agents can understand:

• Questions in plain English - No need for specific commands
• Context and follow-up questions - They remember the conversation
• Intent and meaning - They understand what users really want
• Variations in language - Different ways of asking the same question

​

Knowledge Integration

Your agents use your knowledge base to:

• Find relevant information - They search through your documents and videos
• Provide accurate answers - They use your specific business information
• Give context-appropriate responses - They understand your business context
• Stay up-to-date - They use the latest information you’ve provided

​

Response Generation

Your agents provide:

• Helpful and accurate responses - Based on your knowledge and instructions
• Appropriate tone and style - Matching your brand and business personality
• Complete information - They provide thorough answers, not just brief responses
• Follow-up suggestions - They can suggest next steps or related topics

​

Agent Personality and Voice

​

Customizing Agent Personality

You can customize your agents to match your brand: Tone and Style:

• Professional and formal
• Friendly and casual
• Helpful and patient
• Enthusiastic and energetic

Communication Style:

• Direct and to-the-point
• Detailed and comprehensive
• Conversational and engaging
• Technical and precise

​

Agent Voice Options

Your agents can have different voices: Female Voice:

• Professional and friendly
• Clear and easy to understand
• Warm and approachable
• Great for customer support

Male Voice:

• Professional and authoritative
• Clear and confident
• Reliable and trustworthy
• Great for technical support

​

Agent Knowledge and Learning

​

How Agents Use Knowledge

Your agents access your knowledge base to:

1. Search for relevant information when users ask questions
2. Find the best matches based on content relevance
3. Generate responses using your specific business information
4. Provide accurate answers with proper context

​

Knowledge Types

Your agents can use: Document Knowledge:

• PDF files with product information
• Word documents with policies and procedures
• Text files with FAQs and guides

Video Knowledge:

• Tutorial videos showing how to do things
• Product demonstration videos
• Training content and explanations

Text Knowledge:

• Direct text input about your business
• Notes and reminders
• Custom instructions and information

​

Continuous Learning

Your agents improve over time by:

• Learning from interactions - They understand what users need
• Refining responses - They get better at providing helpful answers
• Adapting to your business - They become more familiar with your context
• Improving accuracy - They provide more precise and relevant information
• Interface exploration - Through simulations, they learn your actual application interface

​

Simulation Learning

One of Marketrix’s most powerful features is simulation learning. Your agents can:

• Explore your actual application through automated simulations
• Learn real interface elements like button names, locations, and workflows
• Understand user journeys by following actual paths through your product
• Stay current with your application as it evolves
• Provide specific guidance based on real interface experience

This hands-on learning makes your agents significantly more effective than those that only rely on documentation. Learn more about simulations and how they transform agent capabilities.

​

Agent Performance and Monitoring

​

Performance Metrics

You can monitor your agents’ performance: Response Quality:

• Accuracy of answers
• Helpfulness of responses
• User satisfaction ratings
• Resolution rates

Usage Statistics:

• Number of conversations
• Most common questions
• Response times
• User engagement

Knowledge Usage:

• Which knowledge is accessed most
• Search success rates
• Content effectiveness
• Knowledge gaps

​

Improving Agent Performance

Based on performance data:

• Update knowledge - Add missing information
• Refine instructions - Improve how agents behave
• Add more content - Fill gaps in knowledge
• Test and iterate - Continuously improve

​

Common Use Cases

​

E-commerce Support

Product Questions:

• “What are the features of this product?”
• “How do I choose the right size?”
• “What’s included with this purchase?”

Order Support:

• “How do I track my order?”
• “Can I change my shipping address?”
• “What’s your return policy?”

Technical Support:

• “How do I set up this product?”
• “Why isn’t this working?”
• “Can you help me troubleshoot?”

​

SaaS Application Support

Feature Help:

• “How do I use this feature?”
• “What does this setting do?”
• “Can you show me how to set this up?”

Account Management:

• “How do I change my password?”
• “Can I upgrade my plan?”
• “How do I add team members?”

Integration Support:

• “How do I connect this to my other tools?”
• “What APIs are available?”
• “Can you help me set up the integration?”

​

Service Business Support

Service Information:

• “What services do you offer?”
• “How much does this cost?”
• “When are you available?”

Appointment Scheduling:

• “How do I book an appointment?”
• “Can I reschedule my appointment?”
• “What should I bring to my appointment?”

General Questions:

• “Where are you located?”
• “What are your business hours?”
• “How can I contact you?”

​

Getting Started with Agents

Ready to create your first agent? Here’s what to do next:

​

Agent Best Practices

​

Creating Effective Agents

Start with clear purpose:

• Define what your agent should help with
• Identify your target audience
• Set clear expectations for behavior

Provide comprehensive knowledge:

• Upload relevant documents and videos
• Include common questions and answers
• Keep information up-to-date

Test and iterate:

• Test your agent thoroughly
• Gather feedback from users
• Continuously improve based on results

​

Maintaining Agent Quality

Regular updates:

• Update knowledge when information changes
• Refine instructions based on performance
• Add new content as needed

Performance monitoring:

• Track response quality and accuracy
• Monitor user satisfaction
• Identify areas for improvement

​

Next Steps

Now that you understand agents, you’re ready to:

1. Create your first agent - Build an AI assistant for your business
2. Write effective instructions - Tell your agent how to behave
3. Assign knowledge - Give your agent access to your information
4. Test in the playground - Make sure your agent works correctly

Your AI agents are the key to providing amazing customer support. With the right setup, they can handle most user questions and provide helpful, accurate responses 24/7.

​

Getting Help

If you need assistance with agents:

• Check the troubleshooting guide for common issues
• Review agent instructions for clarity and completeness
• Test your agents in the playground
• Contact support for additional help

You’re now ready to create intelligent AI agents that will transform your customer support! 🚀