Skip to main content

Topics

Topics help you understand what your customers are actually talking about. Instead of manually categorizing conversations, botBrains uses AI to automatically cluster similar conversations together, revealing patterns in customer needs, pain points, and requests.

Why Topics Matter

Understanding conversation topics gives you actionable insights to improve your product and service:
  • Identify Common Issues: Discover which problems come up most frequently
  • Spot Emerging Trends: Catch new topics before they become major issues
  • Prioritize Resources: Focus your team on the topics that matter most
  • Measure Impact: Track how topic volumes change after product updates or changes
  • Improve AI Performance: Understand knowledge gaps and add targeted content
Instead of reading thousands of conversations manually, topics surface patterns automatically, saving you hours while ensuring nothing important gets missed.

How Topic Analysis Works

botBrains uses advanced AI clustering to automatically categorize your conversations:

Automatic Topic Detection

  1. Initial Setup: You can start with predefined topics for your industry (e-commerce, SaaS) or provide seed topics to guide the granularity
  2. Nightly Analysis: TopicAI runs every night to detect new topics in your conversations
  3. Daily Categorization: Every conversation is assigned to relevant topics based on its content
  4. Weekly Refinement: New topics are identified each Monday to capture emerging patterns

Topic Structure

Each topic has:
  • Title: A clear, descriptive name (e.g., “Shipping Delays”, “Refund Requests”)
  • Description: Detailed explanation of what the topic covers
  • Conversation Count: How many conversations match this topic
  • Trend Indicator: Whether the topic is growing, declining, or stable

Predefined Topic Templates

botBrains offers industry-specific topic templates to get you started quickly:

E-Commerce Topics

Includes topics like:
  • Product Availability
  • Shipping Costs and Options
  • Delayed Delivery
  • Wrong Item Received
  • Return and Exchange
  • Payment Issues
  • And 20+ more

SaaS Topics

Includes topics like:
  • Account Access
  • Billing and Invoices
  • Technical Issues
  • Feature Requests
  • Integration Problems
  • Performance Issues
  • And 10+ more
You can start with these templates and the system will automatically discover additional topics unique to your business.

Using Topics in the Platform

Topics Dashboard

Navigate to Analyze > Topics to access the topics dashboard: Topic Overview Cards
  • Visual cards showing each topic with trend indicators
  • Conversation counts and percentages
  • Quick links to view conversations for each topic
Topic Trends Chart
  • Time-series visualization of topic volumes over time
  • Toggle between absolute numbers and percentages
  • Pie chart showing overall topic distribution
  • Export data for further analysis
Topic Table
  • Sortable list of all topics with detailed metrics
  • Edit topic titles and descriptions
  • View trend badges (New, Growing, Declining, Stable, Inactive)
  • Click through to see all conversations for a specific topic

3D Topic Visualization

Access the interactive 3D visualization to explore topic relationships:
  1. Click Open 3D Visualization on the topics page
  2. Explore the scatter plot where:
    • Each point represents a conversation
    • Colors indicate different topics
    • Proximity shows similarity between conversations
    • Click any point to view that conversation
  3. Use space bar to toggle automatic rotation
  4. Zoom and rotate to explore from different angles
This visualization helps you understand:
  • How well-separated your topics are
  • Which topics are closely related
  • Outlier conversations that don’t fit clear patterns

Filtering by Topics

Use topics throughout the platform to filter data:
  • Conversations: Filter to see all conversations about a specific topic
  • Metrics: Analyze performance metrics by topic
  • Labels: Combine topic filters with labels for deeper analysis

Improving Service with Topics

Identify Knowledge Gaps

If a topic has high volume but low AI resolution rates:
  1. Review conversations in that topic
  2. Identify missing or unclear information
  3. Add targeted knowledge to your data providers
  4. Monitor improvements in resolution rates

Prioritize Product Improvements

Track topic trends over time:
  • Growing topics may indicate new problems or unmet needs
  • Declining topics show where improvements have worked
  • Consistently high-volume topics deserve dedicated resources

Optimize Team Training

Use topics to:
  • Identify which issues require human expertise
  • Create training materials for common topics
  • Update escalation rules based on topic patterns
  • Measure how topic distribution changes as AI improves

Monitor Impact of Changes

After making changes to your product, knowledge base, or AI behavior:
  1. Note which topics should be affected
  2. Monitor topic volumes over the following weeks
  3. Review conversations in affected topics
  4. Validate that changes had the desired impact

Managing Topics

Creating Custom Topics

While topics are detected automatically, you can create custom topics:
  1. Navigate to Analyze > Topics
  2. Click Create Topic
  3. Enter a title and description
  4. Save the topic
The system will then categorize relevant conversations into your custom topic.

Editing Topics

Refine topic titles and descriptions to make them more meaningful:
  1. Find the topic in the topic table
  2. Click the edit icon
  3. Update the title or description
  4. Save changes
Better titles and descriptions help you and your team quickly understand what each topic represents.

Topic Lifecycle

Topics evolve as your business changes:
  • New: Recently discovered topics (less than 7 days old)
  • Growing: Conversation volume increasing by more than 5%
  • Declining: Conversation volume decreasing by more than 5%
  • Stable: Consistent conversation volume (within 5%)
  • Inactive: No recent conversations
Focus on Growing Topics: These often indicate emerging issues that need attention before they escalate.

Best Practices

Start with Seed Topics

When setting up topics:
  • Use industry templates as a starting point
  • Add 5-10 seed topics specific to your business
  • Let the system discover additional topics organically
  • Review and refine monthly

Regular Review Cadence

Establish a routine for reviewing topics:
  • Weekly: Check for new or growing topics
  • Monthly: Refine topic descriptions and clean up inactive topics
  • Quarterly: Analyze long-term trends and adjust strategy

Combine with Other Features

Topics work best when combined with:
  • Labels: Add labels to conversations in specific topics for deeper categorization
  • Metrics: Track resolution rates and response times by topic
  • Audiences: Customize AI behavior for different topics
  • Triggers: Automate workflows based on topic detection

Quality Over Quantity

Don’t create too many topics:
  • Aim for 10-30 meaningful topics rather than 100 overly specific ones
  • Merge related topics if they’re too granular
  • Focus on topics that drive business decisions
Allow Time for Analysis: Topic detection runs nightly, and new topics are identified weekly. Changes to the topic model may take a few days to fully process.

Topic Activation

Topics are powered by TopicAI, which needs to be activated for your project:
  1. Navigate to Analyze > Topics
  2. Click Request TopicAI Activation or Enable Topics
  3. Choose your setup option:
    • Use a predefined template (E-Commerce or SaaS)
    • Provide seed topics
    • Let the system auto-detect without guidance
  4. Wait a few minutes for initial analysis to complete
Once activated, topics will update automatically based on your conversation data.

Troubleshooting

Topics Not Showing Up

If you’re not seeing topics:
  • Verify TopicAI is activated for your project
  • Ensure you have enough conversations (minimum 20-30 recommended)
  • Check that conversations have sufficient content for analysis
  • Wait for nightly analysis to complete

Topics Too Broad or Too Specific

Adjust topic granularity:
  • Provide more specific seed topics to increase granularity
  • Provide fewer, broader seed topics to decrease granularity
  • Edit existing topics to refine their definitions

Conversations in Wrong Topics

The system learns over time:
  • Review misclassified conversations
  • Update topic descriptions to be more specific
  • Create custom topics for unique categories
  • Allow the weekly refinement process to improve accuracy

Next Steps

Now that you understand topics: