Skip to content This library is free, offered as dana. If it helps your work, donate here →
penguin tree ai
← back to libraryDonate here

This is one of 3,622 free prompts in the penguin tree ai library. Sign up once and the whole library opens, with new prompts by email.

open the library →

Discussion guide builder

Builds a complete semi-structured user interview discussion guide from study objectives and research questions — with a timed warm-up, 3-5 core topic sections containing open-ended questions and layered probes, a…

download .md

Purpose

Turn a set of study objectives and research questions into a moderator-ready, semi-structured interview guide — timed, sectioned, and probe-layered — so a researcher can run consistent, bias-resistant sessions and trace every question back to a learning goal.

When to Use

Invoke this skill when the user asks to:

  • Write or structure a discussion guide for user interviews
  • Turn research questions or study objectives into interview questions
  • Prepare a moderator guide for generative, discovery, or exploratory research
  • Build an interview script with warm-up, core topics, and wrap-up
  • Create probes and follow-ups for a semi-structured interview
  • Prep for a usability debrief or contextual inquiry session

Inputs Required

Before starting, collect:

  • Study objectives: 2-4 sentences on what the research must learn and why now
  • Research questions: The 3-6 questions the study is trying to answer (the internal learning goals, not the questions asked to participants)
  • Participant profile: Who is being interviewed (role, context, relationship to the product)
  • Interview length: Total session time in minutes (default: 45 minutes)
  • Study type: Generative/discovery, evaluative/usability, or concept-test
  • Sensitive or off-limits topics: Anything to avoid, or that requires careful phrasing
  • Prototype or artifact: Whether participants will react to a stimulus, and what it is

If study type or interview length is missing, assume a 45-minute generative interview and note the assumption.

Instructions

Step 1: Map Research Questions to Interview Topics

Research questions are internal; participants never hear them directly. Translate each research question into one or more interview topics.

  • Group related research questions into 3-5 thematic core topics
  • For each topic, note which research question(s) it serves — this becomes the moderator cheat sheet later
  • Flag any research question that can only be answered by observation, not self-report, and note it for a task-based moment instead of a direct question

Step 2: Allocate Time Across the Session

Distribute the total interview length across sections using this default split, then adjust:

  • Warm-up: 10-15% of session (build rapport, ease in)
  • Core topics: 70-75% of session (divided across the 3-5 topics by priority)
  • Wrap-up: 10-15% of session (synthesis, gaps, close)

Weight core-topic time by research priority, not equally — the most important research question should get the most minutes.

Step 3: Write the Warm-Up Section

Purpose: build trust, set expectations, and collect light context.

  • Include a plain-language intro line the moderator reads (purpose, recording consent reminder, no-wrong-answers framing, permission to skip)
  • Write 2-4 easy, non-threatening opening questions that surface the participant's context and relationship to the topic
  • Keep these broad and answerable — never lead with the hardest question

Step 4: Build Each Core Topic Section

For each of the 3-5 core topics, produce:

  • Topic title and time estimate
  • Transition line: One sentence the moderator uses to move into this topic naturally
  • 2-4 primary open-ended questions: Non-leading, single-barreled, phrased in the participant's language. Prefer "Tell me about a time...", "Walk me through...", "How do you currently..." over yes/no or hypothetical framings
  • Layered probes under each primary question: 2-3 follow-ups that dig deeper (e.g., "What made that frustrating?", "What did you do next?", "How did you decide?")
  • Optional task or artifact moment: If a prototype is involved, insert a "show and observe" step with a think-aloud prompt

Apply anti-bias rules throughout:

  • No leading questions ("Don't you find X frustrating?")
  • No double-barreled questions (split into two)
  • No jargon the participant wouldn't use
  • Favor past-behavior questions over hypothetical/future-intent questions

Step 5: Write the Wrap-Up Section

Purpose: catch what the structured questions missed and close cleanly.

  • Include a synthesis question ("If you could change one thing about... what would it be?")
  • Include a catch-all ("Is there anything I didn't ask about that feels important?")
  • Include the close: thank-you, next steps, incentive/logistics reminder

Step 6: Produce the Moderator Cheat Sheet

A one-glance reference table mapping each interview topic to the research question(s) it answers, so the moderator knows what "must be covered" versus "nice to have" if time runs short. Mark each topic as Must-cover or Time-permitting.

Output Format

  • Structure: A single moderator-ready guide document with clear section headers and time stamps in the margin/inline
  • Length: 700-1,100 words depending on interview length and topic count
  • Format: Structured document — moderator-read lines in plain text, questions as nested lists (primary question → indented probes), plus one cheat-sheet table at the end
  • Sections: Intro/warm-up → Core Topics 1-N (each with transition, primary questions, probes) → Wrap-up → Moderator Cheat Sheet table
  • Tone: Conversational and human for participant-facing lines; concise and directive for moderator instructions

Quality Checklist

Before delivering output, verify:

  • [ ] Section time allocations sum to the total interview length provided
  • [ ] Every core topic maps to at least one stated research question in the cheat sheet
  • [ ] No primary question is leading, double-barreled, or contains jargon
  • [ ] Each primary question has 2-3 layered probes beneath it
  • [ ] Warm-up opens with easy questions and wrap-up includes a catch-all
  • [ ] Cheat sheet marks each topic as Must-cover or Time-permitting

Notes

  • Semi-structured means the guide is a map, not a script — remind the moderator to follow interesting threads and skip questions already answered organically
  • Past-behavior questions ("Tell me about the last time...") yield far more reliable data than hypotheticals ("Would you use...") — lean on them heavily in generative studies
  • If the study is evaluative/usability, weight core topics toward task-based observation moments and reduce direct-question time
  • Build in slack: a guide that fills 100% of the session leaves no room for the unexpected answers that are usually the most valuable

How to use this skill

This file is a Claude Code skill in Anthropic's standard format — markdown with YAML frontmatter at the top.

Installing in Claude Code:

  • ~/.claude/skills/<folder-name>/SKILL.md — auto-loads in every session anywhere on your machine (recommended for skills you'll use across multiple projects)
  • <your-project>/.claude/skills/<folder-name>/SKILL.md — auto-loads only when working in that project (recommended for project-specific skills)

Restart Claude Code or open a new session to register.

Using in Claude.ai, ChatGPT, Gemini, or other AI tools:

  1. Drag this file into a new chat, or copy and paste its contents
  2. Send your task or request as your next message — the AI will apply the skill where it fits

For Claude Code skill details: https://docs.anthropic.com/en/docs/claude-code/skills