Prompt library
Write a usability test script that does not lead
Scenario-goal phrasing and the label-matching ban are what make a test measure findability instead of reading ability. Scripted neutral probes protect the moment moderators most often contaminate: the silence after a participant gets stuck.
DesignWrite
Last reviewed July 17, 2026
The prompt
Write a usability test script.
What is being tested: {{subject}}
The 3 things we most need to learn: {{questions}}
Produce:
1. Intro script, verbatim: sets the you-are-not-being-tested frame, permission to think aloud, permission to give up, and the recording consent line.
2. Warm-up: one task in the product's general area that calibrates the participant's baseline fluency.
3. Core tasks (3-5): each written as a SCENARIO with a goal ("you want your teammate to see this file"), never as instructions ("click share"). Per task: the success definition, the observation checklist (where they hesitate, what they try first, what they call things), and the do-not-help line the moderator must hold.
4. The probe bank: neutral probes only ("what are you thinking?", "what did you expect to happen?"); list the leading probes moderators reach for under pressure, banned.
5. Wrap: the two closing questions that consistently surface withheld reactions ("what almost made you give up?", "what would you tell a colleague this is for?").
6. The analysis sheet: how observations get coded afterward so 5 sessions produce patterns, not anecdotes.
Rule: if a task name appears in the interface (label-matching), rewrite the scenario until it does not.Run in idaptOpens a new chat with the prompt prefilled. Nothing sends until you press send.
Fill in the variables
| Variable | What it is | Example |
|---|---|---|
| {{subject}} | The prototype or feature | the new file-sharing dialog prototype |
| {{questions}} | What you must learn | do people find permissions; do they trust the link option; is 'workspace' understood |