These prompts cover the stages of moderated usability testing where AI saves the most time: planning test scenarios, analyzing individual session transcripts, synthesizing findings across multiple sessions, and drafting stakeholder-ready reports.
Generate a usability test plan from research questions
I am planning a moderated usability test for [product description]. The test will be conducted [remotely via Zoom / in-person in a lab].
Research questions:
[list 3-5 research questions, e.g., "Can users complete the checkout flow without assistance?"]
Target participants: [description of target users]
Based on these research questions, generate a complete test plan:
1. TASK SCENARIOS: Write 4-6 realistic task scenarios. Each scenario should:
- Set context without using words from the interface
- Have a clear success state
- Be completable in 3-5 minutes
- Cover one research question
2. PRE-TASK QUESTIONS: 3-4 opening questions to build rapport and understand the participant's current behavior in this domain
3. POST-TASK QUESTIONS: For each task, suggest 2-3 follow-up questions to probe comprehension and satisfaction. Include the Single Ease Question (SEQ).
4. POST-SESSION QUESTIONS: 4-5 closing questions about overall experience, comparison to current solution, and willingness to use
5. SUCCESS CRITERIA: For each task, define what constitutes success, partial success, and failure
Format as a ready-to-use moderator guide.Analyze a usability test session transcript
Here is the transcript from a moderated usability test session. The participant was asked to complete [N] tasks on [product description].
Tasks:
[list the tasks and their success criteria]
Transcript:
[paste transcript]
Analyze this session and produce:
1. TASK RESULTS TABLE: For each task, record: completed (yes/partial/no), time taken (if timestamps available), SEQ score (if mentioned), and key observation.
2. USABILITY FINDINGS: List every usability problem the participant encountered. For each:
- What happened (specific behavior)
- Why it happened (what the participant expected vs. what the interface did)
- Severity: Critical (blocked task), Moderate (caused difficulty), Low (minor friction)
- Supporting quote from the transcript
3. POSITIVE OBSERVATIONS: Things that worked well — features the participant found easily, praised, or used correctly on the first try.
4. DESIGN RECOMMENDATIONS: For each usability finding, suggest a specific fix (not just "improve navigation" but "add a breadcrumb trail showing the user's current location within the account settings hierarchy").
5. KEY QUOTES: The 3-5 most revealing quotes from the participant, with context.Synthesize findings across multiple usability test sessions
I have conducted [N] moderated usability test sessions for [product description]. Below are the session summaries:
[paste session summaries — each should include task results, findings, and key quotes]
Synthesize these sessions into a consolidated findings report:
1. FINDINGS TABLE: List each unique usability finding. For each, show:
- Finding description
- How many participants encountered it (e.g., 4/5)
- Severity (Critical / Moderate / Low)
- Impact on which task(s)
2. TOP 5 PRIORITY FINDINGS: Rank by severity × frequency. For each, include:
- The most illustrative participant quote
- A specific design recommendation
3. TASK SUCCESS SUMMARY: Table showing each task, overall success rate, average SEQ score, and the most common failure point.
4. PATTERNS: Are there recurring themes across findings (e.g., "labeling confusion," "missing feedback," "hidden actions")? Group findings into 3-5 theme clusters.
5. WHAT WORKED: Features or flows that all or most participants completed without difficulty — these should be preserved in redesign.
6. NEXT STEPS: Recommend whether to redesign and retest, ship with known issues, or conduct additional research on specific areas.Draft a stakeholder summary from usability test results
I need to present usability test results to [stakeholder audience: e.g., "product managers and engineers who were not present during testing"]. The audience has 15 minutes and prefers concrete evidence over methodological detail.
Here are the consolidated findings from [N] sessions testing [product description]:
[paste findings summary]
Create a stakeholder-ready summary:
1. ONE-PARAGRAPH EXECUTIVE SUMMARY: What was tested, with how many users, and the headline result (e.g., "3 of 5 users could not complete the core checkout task without assistance").
2. TOP 3 FINDINGS: Each in this format:
- Problem (one sentence)
- Evidence (participant count + one quote)
- Recommendation (specific action)
- Business impact if not fixed (e.g., "users who cannot find the return policy are likely to contact support, increasing support volume")
3. WHAT WORKS WELL: 2-3 positive findings to balance the report
4. RECOMMENDED NEXT STEPS: What the team should do now (fix and retest, ship with monitoring, etc.)
Keep the total length under 500 words. No jargon — stakeholders care about what's broken, what it costs them, and what to do about it.