Skip to main content

​
Key Concepts Explained

Personas: AI-generated user profiles that represent different types of people from your interviews. Think of them as fictional characters based on real data — each persona captures the behaviors, motivations, and characteristics of a group of similar participants. This helps you understand your audience in digestible segments rather than trying to process dozens of individual responses. Each persona:
  • Represents a cluster of real participants
  • Aggregates shared behavioral and attitudinal patterns
  • Includes supporting quotes from actual transcripts
  • Is fully traceable back to source interviews
Classification: The process where AI automatically sorts your interview participants into personas based on their responses. Instead of manually reading through every transcript to find patterns, the system identifies common traits and groups similar people together under each persona. Participants are assigned based on similarity across:
  • Themes and coded responses
  • Emotional reactions
  • Language patterns
  • Behavioral signals
  • Demographic indicators
Persona Chat: An interactive feature that lets you have conversations with your personas as if they were real people. The responses come from actual interview data, so when “Amanda” answers your questions, she’s drawing from real quotes and insights from participants who match her profile. Persona Chat synthesizes patterns across that persona’s participant group — it does not invent new data.

​
Pro Tips

  • Start with quality data: The more detailed and varied your interview transcripts, the richer and more accurate your personas will be. Aim for at least 10-15 interviews before generating personas.
  • Use both approaches strategically: If you have existing personas from other research, start with those to see how your current data maps to known user types. Use auto-generation when exploring new markets or validating assumptions.
  • Leverage the chat feature for presentations: Instead of showing raw data in meetings, have a conversation with a persona. Ask “Lucy” about feature preferences or pain points to make insights more memorable and relatable.
  • Cross-reference with quotes: Always click through to the actual transcript quotes that support persona characteristics. This helps you understand the “why” behind each trait and adds credibility to your findings.

​
Quick Reference

Two Setup Methods:
  • Import existing personas → AI classifies interviews into your categories
  • Auto-generate new personas → AI creates personas from scratch based on your data
Key Persona Information:
  • Name and basic demographics
  • Job/role characteristics
  • Behavioral patterns and preferences
  • Supporting quotes from actual interviews
  • Interview classification (which participants match this persona)
Chat Features:
  • Ask open-ended questions about preferences, pain points, motivations
  • Upload images or videos for persona reactions
  • Get responses backed by real interview data
  • Save conversations for team sharing

​
Complete Written Guide

If you prefer to read or want a reference, here’s the complete step-by-step process:

​
Step 1: Choose Your Persona Method

Option A: Import Existing Personas (not standard enabled, available upon request)
  1. Navigate to the Analysis section of your study
  2. Select “Personas” from the analysis options
  3. Choose “Import existing personas”
  4. Upload or input your current persona definitions
  5. The AI will classify your interview participants into these predefined categories
Option B: Auto-Generate New Personas
  1. Go to Analysis → Personas
  2. Select “Auto-generate personas”
  3. The system will analyze all your transcripts and create persona profiles automatically
  4. This process takes several minutes depending on your data volume

​
Step 2: Review Generated Personas

Once processing is complete, you’ll see:
  • Persona names: AI-generated names for each user type
  • Key characteristics: Job roles, behaviors, and defining traits
  • Classification results: Which interviews belong to each persona
  • Supporting quotes: Real participant responses that define each persona

​
Step 3: Explore Individual Personas

For each persona, you can:
  1. Read the profile: Review the AI-generated characteristics and traits
  2. View classified interviews: See which participants match this persona
  3. Browse quotes: Read actual responses that support the persona definition
  4. Click quote sources: Jump directly to the original transcript location

​
Step 4: Use Persona Chat

  1. Start a conversation: Click on any persona to open the chat interface
  2. Ask questions: Type questions about preferences, behaviors, or opinions
  3. Upload media: Share images, videos, or documents for persona reactions
  4. Save insights: Important responses can be saved or shared with your team

​
Best Practices

Before You Start:
  • Ensure you have sufficient interview data (minimum 10-15 transcripts recommended)
  • Clean up transcript quality by removing filler words and unclear sections
  • Consider your research goals - are you validating existing assumptions or discovering new user types?
While Working:
  • Review persona characteristics for accuracy and distinctiveness
  • Check that each persona represents a meaningful segment of your audience
  • Test the chat feature with various question types to understand persona depth
Quality Control:
  • Verify that supporting quotes actually match the persona characteristics
  • Ensure personas are different enough from each other to be useful
  • Cross-reference chat responses with original transcript data for accuracy

​
Important Notes

  • Personas represent patterns — not individual participants
  • They synthesize qualitative data and are not statistically weighted segments
  • Smaller clusters may be directional
  • Persona composition may change if regenerated after adding new interviews
Anything missing? Let us know at support@conveo.ai and we’ll help you out!