HomeCasa Heuristic evaluation of a screen or flow using Nielsen’s 10 heuristics
Evaluate a screen or flow against Nielsen’s 10 usability heuristics: findings with 0-4 severity, evidence, recommendations, effort, priorities and quick wins. Designed for ChatGPT, Claude, Gemini and Microsoft Copilot. It includes 5 customizable variables and you can copy it for free in one click.
Intermediate · Text · Pending editorial review
Prompt
Act as a usability expert who runs heuristic evaluations for product teams.
Context:
- Product: {{product}}
- Screen or flow to evaluate (detailed description, visible copy and steps): {{flow}}
- User goal in this flow: {{goal}}
- User type: {{user}}
- Platform: {{platform}}
Deliverables:
1. A summary of how you understand the flow and the assumptions you make; if there are important gaps, ask me before evaluating.
2. An evaluation against Nielsen's 10 heuristics (visibility of system status, match with the real world, user control and freedom, consistency and standards, error prevention, recognition rather than recall, flexibility and efficiency, aesthetic and minimalist design, helping users recognize and recover from errors, help and documentation).
3. A findings table: ID, heuristic, location (step or element), issue, evidence from the description, severity from 0 to 4 on Nielsen's scale, a concrete recommendation and estimated effort (low, medium, high).
4. Strengths of the design worth keeping.
5. The top 5 issues, ranked by severity and by how often a user would hit them.
6. Quick wins that take less than a day versus deeper changes.
7. What a heuristic evaluation cannot confirm and what to validate with real users.
Constraints:
- Base every finding on what appears in the description; if something cannot be verified, mark it "to verify".
- Do not invent metrics, usage data or test results.
- Be specific: every recommendation should be easy to turn into a ticket.
Format: a short summary, the findings table, a list of priorities and quick wins, and a final limitations section.How to use it
- Describe the screen or flow step by step, including the copy users see. 2. Review the table and drop any "to verify" findings that do not apply. 3. Turn the top 5 into tickets and validate the changes with users.
Usage example
Product: medical appointment app · Flow: booking a specialist in 4 steps · Goal: book the first open slot · User: patient over 60.
Sample output
Summary
A 4-step flow to book the first available dermatology appointment. I'm assuming there's no "first available slot" option, since it isn't in the description.
Findings
| ID | Heuristic | Location | Issue | Severity (0-4) | Recommendation | Effort |
|---|---|---|---|---|---|---|
| H1 | Recognition rather than recall | Step 3 | The calendar shows no open slots until you tap each day | 4 | Mark days that have slots and disable the rest with "No appointments" text | Medium |
| H2 | Flexibility and efficiency | Steps 1-3 | No shortcut for "first available", which is exactly the goal | 3 | A "First available appointment" button after choosing a specialty | Medium |
| H3 | Match with the real world | Step 4 | "Submit" doesn't say an appointment is being confirmed | 2 | Change to "Confirm Tuesday 10:00 AM appointment" | Low |
| H4 | Aesthetic and minimalist design | Step 1 | 30 specialties with no search or ordering | 3 | Add search and put the most common specialties first (to verify with real data) | Low |
Priorities
- H1 · 2. H2 · 3. H4 · 4. H3…
Quick wins (under a day)
- Rewrite the button label (H3).
- Add search to the list (H4).
Limitations
This evaluation can't confirm how many patients over 60 drop off at step 3; that needs testing with real users…
Key facts
- Category
- UI/UX design · Prototypes
- Output type
- Text
- Level
- Intermediate
- Designed for
- ChatGPT, Claude, Gemini and Microsoft Copilot
- Verified in
- Pending editorial verification
- Variables
- Product, Screen or flow, User goal, User type and Platform
- Language
- English
- Price
- Free, no sign-up
- Updated
- Nov 7, 2026
Frequently asked questions
Can AI do a reliable heuristic evaluation?
It can give you a useful, structured first pass from a good description or screenshots, but it does not replace several expert evaluators or user testing. Use it to prioritize and prepare validation.
What does the 0-4 severity scale mean?
0 means not a usability problem, 1 cosmetic, 2 minor, 3 major and 4 catastrophic, which must be fixed before launch. It combines how often the problem occurs, its impact and its persistence.
What is this prompt for?
Evaluate a screen or flow against Nielsen’s 10 usability heuristics: findings with 0-4 severity, evidence, recommendations, effort, priorities and quick wins.
Which AI tools does it work with?
It is designed for ChatGPT, Claude, Gemini and Microsoft Copilot. We have not verified it in these tools yet, so results may vary between them.
How do I customize the prompt?
Click “Customize” and fill in the variables (Product, Screen or flow, User goal, User type and Platform). The text updates instantly and you can then copy it in one click. Values are stored only in your browser.
Do I need experience to use it?
It is intermediate: it helps to know the topic so you can provide good context and review the result critically.
Is this prompt free?
Yes. Every HomeCasa Prompts prompt is free, with no sign-up or subscription. You can also use them in professional projects.
Can I trust the AI output?
Treat it as a high-quality draft: always check data, figures and important claims before publishing or making decisions.