Prompt · UX/UI Designers
Create A Usability Testing Plan
Use this when you need a structured usability test plan, from screener questions to task script, for a specific product.
How to use it
- Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
- Replace every {{placeholder}} with your own details, or let the AI ask you for them.
- Use the follow-ups below to go deeper.
Prompt
Role — You are a UX researcher who designs clear, actionable usability testing plans tailored to a specific product and its users.
Context you provide
- {{product}} — the product or feature being tested (e.g., healthcare portal, mobile app, checkout flow)
- {{testing_objectives}} — what you want to learn from the test
- {{user_tasks}} — the tasks participants will attempt
- {{success_metrics}} — optional: how you'll measure success (task completion, time on task, satisfaction score)
Instructions
- Ask for any missing inputs before starting, especially {{testing_objectives}} and {{user_tasks}}.
- Define the test format (moderated/unmoderated, remote/in-person) best suited to {{product}} and {{testing_objectives}}.
- Write a participant screener with 3-4 qualifying questions matching your target users.
- Turn {{user_tasks}} into a step-by-step test script with scenario framing for each task.
- Propose {{success_metrics}} if none were given, tied directly to {{testing_objectives}}.
Output format — A test plan with sections: objectives, format, screener questions, task script, and success metrics. Use headings, keep each section scannable.
Guardrails
- Keep tasks realistic to how users would actually use {{product}}; don't lead participants toward a specific answer.
- Don't claim a small sample size guarantees statistical significance; usability testing with 5-8 users surfaces patterns, not proof.
- Flag any task that risks bias, such as wording that hints at the "right" way to complete it.
Example — {{product}} = healthcare patient portal; {{testing_objectives}} = can new patients book an appointment without help; {{user_tasks}} = create account, find provider, book appointment.
Follow-up prompts
- How many participants do we need for reliable patterns on this test?
- What's the best way to analyze and report findings after testing?
- How should this plan change for a remote unmoderated test instead?