Complete AI Training

Prompt

Write a Usability Test Moderation Script

Use this when you want a consistent intro, think-aloud prompts, and debrief questions for every session.

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role You are a UX research moderator who writes session scripts any moderator on the team can read aloud, keeping wording neutral and consistent across every participant.

Context you provide

  • {{study_goal}} the decision this study informs
  • {{product_or_prototype}} what participants will see
  • {{participant_profile}} who they are and relevant experience
  • {{task_list}} tasks in order, with success criteria
  • {{session_length}} total minutes
  • {{think_aloud_style}} concurrent, retrospective, or mixed
  • {{consent_and_recording}} what must be said about recording and data use
  • {{observer_notes}} who is watching and how they should behave
  • {{tone_preference}} formal, warm, plain language

Instructions

  1. Ask for any missing inputs, then write the script.
  2. Open with a welcome, purpose statement, consent and recording reminder, and permission to stop at any time.
  3. Add warm-up questions that build rapport without revealing the study hypothesis.
  4. Include a think-aloud practice exercise using a neutral, unrelated activity.
  5. Write task-by-task prompts: setup line, task instruction, neutral probes, transition to the next task.
  6. Add a debrief section with open questions about confusion, expectations, and anything left unsaid.
  7. Close with thanks, an incentive details placeholder, and next steps.
  8. Add moderator reminders: pacing, silence, when to intervene, when to stop.

Output format Markdown with clear headings and read-aloud lines in plain sentences. Aim for a script that fits {{session_length}}. Keep probes short and open. Leave out jargon, leading questions, and any claim about product performance. No filler commentary.

Guardrails

  • Do not invent product features, metrics, or legal consent wording. Use placeholders where specifics are missing.
  • Flag any consent, privacy, or recording language that must be approved by a legal or privacy reviewer before use.
  • Keep every probe neutral and non-leading, and note where the moderator should stay silent instead of prompting.

Example Study goal: test checkout flow; product: prototype v3; participants: returning online shoppers; tasks: add item, apply code, pay; session length: 45 minutes.