# Usability Test Study Designer

Create a focused usability study with unbiased tasks, observation criteria, session structure, analysis rules, and ethical participant handling.

## Prompt

You are a UX research lead who specializes in task-based usability testing and actionable product evidence.

Inputs:
1. Product, prototype stage, and research objective: {{study_context}}
2. Target participants and recruitment constraints: {{participants}}
3. Critical tasks, flows, and design questions: {{research_questions}}
4. Session format, tools, timeline, and team capacity: {{operations}}
5. Privacy, consent, accessibility, and reporting constraints: {{constraints}}

Do the following:
1. Convert broad design questions into observable research questions and define what this study can and cannot answer.
2. Recommend participant segments and sample logic based on behavior and experience rather than unsupported demographic assumptions.
3. Write realistic, non-leading task scenarios that give participants a goal without revealing interface labels or the intended path.
4. Define moderator prompts, observation fields, success and recovery criteria, severity rules, timing guidance, and a note-taking structure that separates behavior, quotation, and interpretation.
5. Produce a research plan, screener criteria, consent topics, session script, task set, analysis framework, reporting template, and pilot checklist. Avoid using completion rate from a small qualitative study as population-level proof.

## Best for

Product and design teams planning moderated or unmoderated usability tests that need reliable behavioral evidence rather than opinion gathering.

## Compatible tools

- Claude
- ChatGPT

## How to use

- Define the decisions the study should inform.
- Recruit by relevant behavior and experience.
- Pilot every task before participant sessions.
- Keep research participation separate from performance review.

## Customization tips

- Avoid interface words in task instructions.
- Record recovery behavior, not only success.
- Include accessibility needs in recruitment.
- Separate observations from interpretations during analysis.

## Example input

Product: Tablet app used by hospital porters to accept and complete transport requests. Stage: clickable prototype. Questions: Can porters identify urgent requests, report a delay, and recover from accepting the wrong job? Participants: porters with less than three months or more than two years of experience. Operations: six moderated 45-minute sessions in one week. Constraints: no patient data, voluntary participation through staff council, screen-reader compatibility included, and sessions cannot affect employment assessment.

## Example output

The plan defines observable questions around priority recognition, delay reporting, and mistaken acceptance recovery. Scenarios use fictional wards and ask participants to complete work goals without naming buttons. The sample includes three experience levels where feasible and at least one assistive-technology user through accessible recruitment. Notes separate actions, statements, and analyst inference. Severity combines task impact, recovery, frequency across sessions, and safety relevance rather than counting errors alone.