Setting Up Your Topic Guide

​

Key Concepts Explained

AI Study Design Assistant: Think of this as your research partner. It helps you build your study by suggesting questions, objectives, and improvements based on your research goals. You’re always in control - you can accept, decline, or modify any suggestions. AI Check Quality Assistant: Think of this as your quality reviewer. It helps you assess your study by checking for consistency, clarity, and opportunities for improvement across objectives, questions, and overall design. You stay in control—you can review, accept, or adjust any suggestions it provides. Topic Guide Structure: Your guide has four main parts:

• Research Briefing: The overall context and goals
• Research Objectives: What you want to learn (usually 1–3 objectives). These are analytical lenses — they drive themes, summaries, and insights, but don’t contain questions themselves.
• Sections: Structural groupings of questions in the interview. Use a new section when a group of questions needs a shared stimulus, shared randomization, or shared routing.
• Questions: The specific questions inside each section.

See Sections for a full walkthrough of how sections and research objectives interact. Studies created before sections were introduced use an older model where research objectives directly contained questions; that model is still supported for those studies.

​

Pro Tips

• Start with context: The more background you give the AI, the better its suggestions will be
• Use the prompt box: You can ask for anything - “make this question multi-select” or “add more follow-ups”
• Follow-up settings matter: This controls how much the AI will probe after each question. More follow-ups = deeper insights but longer interviews
• Context box is private: What you write here guides the AI but isn’t shown to participants
• Leverage the Quality Control AI to get smart recommendations to improve your research objectives and questions—and easily update them by accepting the suggested changes

​

Quick Reference

Question Types:

• Open-ended: Best for exploratory research
• Single select: When you want one clear answer
• Multi-select: When multiple answers are possible

Follow-up Levels:

• Automatic probing: Automatic probing lets the AI moderator naturally decide when and how to ask follow-up questions, adapting to each participant’s responses while staying guided by your objectives and probe instructions.
• No follow-up questions: no additional quesitons on top of the main question
• 1-2 follow-ups
• 2+ follow-ups

​

Language Settings

Conveo supports multilingual studies, allowing you to run research with participants in their native language while receiving your reports in your preferred language. Language settings are configured at the study level, before launch. There are two independent language settings:

• Participant Language: The language your participants will interact in: all questions, prompts, and AI moderator responses will be delivered in this language
• Reporting Language: The language used for your analysis, transcripts, and AI-generated reports — independent of what language participants spoke in

View all 51 supported participant languages

Languages
Afrikaans	Estonian	Korean
Arabic	Finnish	Latvian
Bosnian	French	Lithuanian
Bulgarian	French (Canada)	Macedonian
Chinese (Hong Kong)	German	Malay
Chinese (Simplified)	Greek	Norwegian
Chinese (Traditional)	Hindi	Polish
Croatian	Hungarian	Portuguese (Brazil)
Czech	Icelandic	Portuguese (Portugal)
Danish	Indonesian	Romanian
Dutch (Belgium)	Italian	Russian
Dutch (Netherlands)	Japanese	Serbian
English (Australia)	Slovak	Slovenian
English (United Kingdom)	Spanish	Spanish (Mexico)
English (United States)	Swedish	Tagalog
Tamil	Thai	Turkish
Ukrainian	Urdu (Pakistan)	Vietnamese

View all 8 supported reporting languages

Languages
English (United States)
German
Spanish
French
Italian
Dutch (Netherlands)
Polish
Portuguese (Portugal)

---

​

Stimuli Video Controls

Use Prevent skipping ahead when answer quality depends on full exposure to a video stimulus (for example ad tests, concept films, or product demo walk-throughs). It helps reduce bias from partial viewing. When enabled:

• Participants can pause and replay, but cannot seek forward
• The interview stays blocked until the uploaded video reaches the end
• It works for single uploaded videos and uploaded video assets inside randomized groups

When to leave it off:

• Exploratory studies where natural viewing behavior is more important than strict control
• Long videos where strict gating may increase drop-off risk

Limitations:

• It is supported for uploaded video stimuli
• It is not supported for YouTube video stimuli
• In randomized stimulus groups, it applies only to uploaded video assets in that group
• If strict control is required, upload the video file instead of using a YouTube link

---

​

Complete Written Guide

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

​

Step 1: Access the Dashboard

1. Log into your Conveo account
2. You’ll see the main dashboard with:
   • Prompt box at the top for communicating with the AI
   • Study summary at the bottom showing your drafted and completed studies

​

Step 2: Start Your Topic Guide

1. Click the prompt box and describe what you want to research
2. You can provide context in two ways:
   • Upload documents (PDFs, Word docs, etc.)
   • Type directly in the text area
3. Click “Send” to start working with the AI design assistant

​

Step 3: Review Research Briefing

The AI will automatically generate:

• Study title based on your research description
• Research briefing with overall context and goals
• Review and accept/decline these suggestions as needed

​

Step 4: Set Research Objectives

Research objectives are the analytical lenses for your study. They define what you want to learn, and drive how themes, summaries, and insights are generated.

1. The AI will suggest research objectives based on your briefing
2. You can:
   • Accept all suggestions
   • Accept some and decline others
   • Manually edit any objective
3. Typically aim for 1–3 clear, focused objectives
4. Research objectives don’t contain questions — questions live in sections (see next step)

​

Step 5: Build Your Sections and Questions

Sections are the structural containers for your questions in the interview. Every question lives inside a section.

1. Start with the default section and add questions to it
2. Create an additional section when a group of questions needs to:
   • Share a stimulus (image or video shown across the group)
   • Be randomized together to reduce order bias
   • Be routed together based on a screener answer
3. For each question, the AI can generate relevant content based on your objectives
4. You can:
   • Use the prompt box to ask for specific types of questions
   • Accept/decline individual questions
   • Manually edit questions as needed
   • Drag questions between sections in the Topic guide

See Sections for the full reference.

​

Step 6: Configure Question Settings

For each question, you can set: Question Type:

• Open-ended: Participants provide free-form answers
• Single select: Multiple choice with one answer allowed
• Multi-select: Multiple choice with multiple answers allowed

Stimuli (optional):

• Add images, videos, or Figma designs
• Include links for UX testing
• Upload documents for reference
• For video stimuli, Prevent skipping ahead works for uploaded videos only (not YouTube videos)

Follow-up Configuration:

• Automatic: Let the AI decide based on context
• No follow-up questions: No additional questions on top of the main question
• 1-2 follow-ups
• 2+ follow-ups

Context Box (important):

• Write background information for the AI moderator
• This guides how the AI will probe and follow up
• Note: This is private and not shown to participants

Routing (optional):

• Apply routing conditions to main questions
• Refer to the “Question routing” tab for detailled guide

​

Step 7: Fine-tune Your Guide

1. Use the prompt box to make changes:
   • “Make question 3 multi-select”
   • “Add more follow-ups to the pricing section”
   • “Change the study title to be more specific”
2. The AI will propose changes that you can accept or decline

​

Step 8: Configure Study Settings

1. Language Settings:
   • Set participant-facing languages
   • Choose reporting language
2. Study Presentation:
   • Add a study title participants will see
   • Write an introduction brief for participants

​

Best Practices

Before You Start:

• Have a clear research goal in mind
• Gather any relevant documents or materials
• Think about your target participants

While Building:

• Start with broad context, then get specific
• Use the prompt box liberally - the AI can handle complex requests
• Don’t worry about perfection - you can always edit later

Question Design:

• Mix question types for variety
• Use open-ended questions for exploratory research
• Use single/multi-select for specific feedback
• Set follow-up levels based on how deep you want to go

Quality Control:

• Test your topic guide with colleagues before launching to catch unclear questions
• Leverage the quality control AI recommendations to pressure test the questions and objectives
• Review the AI’s question suggestions critically - accept what works, modify what doesn’t
• Ensure your research objectives align with your overall business or research goals
• Check that follow-up settings match your time constraints and depth requirements

---

Anything missing? Let us know at support@conveo.ai and we’ll help you out!