Chat 2.0

​

Introduction

• If you ask a quantitative question, it uses quant analysis tools.
• If you ask a conceptual question, it uses concept testing tools.
• If you ask a thematic question, it uses thematic analysis tools.

​

How to Use Chat 2.0

1. Ask a Question
   • Be clear and specific. Chat 2.0 will select the right analysis approach based on your question.
   • For complex, multi-step research queries, the Chat 2.0 will create a to-do list showing the steps it will take and ask you to confirm before proceeding.
2. Review Artefacts
   • Chat 2.0 outputs artefacts—the answers to your questions.
   • Every artefact includes citations, so you can check the supporting evidence.
3. Use in Your Workflow
   • Incorporate artefacts into your analysis and reporting.
   • Always review citations to ensure the findings are well supported.

​

Best Practice Recommendations

• Use Chat 2.0 when you need comprehensive synthesis, quant comparisons and adaptive responses.
• Use Chat 1.0 only for lightweight exploratory work or when you want a quick scan without structure.
• Always review citations before using artefacts in reporting.
• Think of Chat 2.0 as a partner in structured analysis—you guide with the questions, and it provides systematic evidence-based outputs.

​

Examples of Prompts:

• Thematic breakdowns
  • “Summarise the key themes in the data.”
  • “Identify outliers in the data.”
• Participant-level comparisons
  • “Find the similarities and differences between the Female and Male participants. Run two separate analyses.”
• Concept and stimulus reactions
  • ”What did participants think about Concept A?”
  • ”Breakdown perception of Concept A for segments X and Y”
• Quantitative counts
  • “How many participants mentioned XX in their response?”
  • “How many people knew about YY?”
• Journey mapping
  • “Analyse participants’ decision-making in XX?”
  • “What part of the process did participants find the most difficult?”
• Segment-based analysis
  • “What are the key pain points per segment?”
  • “What segments can you find in the data, aside from the ones I’ve already labelled?”
• Evidence retrieval
  • “Find evidence supporting the hypothesis that participants thought XX about YY.”
• Quote Retrieval
  • “Give me 2 quotes from every participant in segment Y”
  • “Give me 1 quote from every female participant about what they liked about concept A”
• Categorisation Question
  • “Break down the participants into part time and full time workers”
  • “Break down the participants by the days of the week they work”
• Comparative Analysis
  • “Compare and contract Segment A and Segment B on customer satisfaction”
  • “What do engineers and designers have in common regarding work-life balance”

​

Additional Capabilities

• Example Prompt: “What are the top 5 themes related to participant voting behaviours?”

• Example Prompt: “How many participants said their families influenced their voting decisions?”

• Example Prompt: “Conduct concept analysis on #Novartis”

• Example Prompt:
  • Pie Chart (mutually exclusive): “Break down participants by part time and full time workers”
  • Bar Chart (not mutually exclusive): “Break down the days of the week participants worked”

• Example prompts:
  • For Similarities: “Find similarities between Segment A and Segment B on customer satisfaction”
  • For Differences: “What are the differences between power users and casual users regarding feature requests?”
  • For Both: “Compare and contrast Segment A and Segment B on product usability”

​

Filtering in the Chat