Heuristic Evaluation for Fashion GPT on Myntra — a UX case study

I performed heuristic evaluation for the FashionGPT feature on the Myntra app using Nielsen’s 10 general principles for interaction design. Myntra is India’s leading online shopping application.

A heuristic evaluation is a usability inspection method mainly used to identify any design issues associated with the user interface.

There are 10 general principles by Jakob Nielsen’s for interaction design. They are called “heuristics” because they are broad rules of thumb and not specific usability guidelines.

Let’s see the evaluation of the FashionGPT feature on the Myntra app according to the 10 principles.

1. Visibility Of System Status

The Fashion GPT feature should provide clear and immediate feedback to users regarding the system’s status and responses. Users should be aware of any ongoing processes or delays.

Evaluation Findings:

• The Fashion GPT feature lacks loading indicators or progress bars, making it unclear to users whether the system is processing their requests or if there are delays.

Proposed Solution:

• Implement loading indicators or progress bars to provide visual cues that indicate the system’s status and processing time, ensuring users are aware of the ongoing processes.

2. Match between System and Real World

The Fashion GPT feature should use language, concepts, and representations familiar to users, aligning with their mental models and expectations.

Evaluation Findings:

• The Fashion GPT feature may use technical jargon or complex terminology that is not easily understandable to users, resulting in a mismatch between the system and users’ mental models.

Proposed Solution:

• Use user-friendly language, familiar concepts, and easily understandable representations within the Fashion GPT feature to align with users’ mental models and improve comprehension.

3. User Control and Freedom

Users should have control over the Fashion GPT feature and the freedom to navigate and explore without fear of making irreversible errors.

Evaluation Findings:

• The Fashion GPT feature lacks options for users to customize or adjust the behavior of the model, limiting user control and personalization.
• There are no clear undo/redo options or a way to revert or correct unintended inputs or actions within the Fashion GPT feature.

Proposed Solution:

• Provide customization options that allow users to adjust the Fashion GPT model’s outputs based on their preferences and style.
• Implement undo/redo functionalities or mechanisms that allow users to easily revert or correct inputs or actions within the Fashion GPT feature.

4. Consistency and Standards

The Fashion GPT feature should follow consistent design patterns, interactions, and visual elements within the Myntra app, ensuring a coherent and familiar user experience.

Evaluation Findings:

• The design patterns, placement of elements, and visual cues within the Fashion GPT feature may differ from the rest of the Myntra app, leading to inconsistencies and confusion.

Proposed Solution:

• Ensure that the Fashion GPT feature aligns with the overall design patterns, interactions, and visual elements of the Myntra app, providing a consistent and seamless user experience.

5. Error Prevention

The Fashion GPT feature should anticipate and prevent errors by providing clear instructions, constraints, and error prevention mechanisms.

Evaluation Findings:

• The Fashion GPT feature may not provide clear instructions or constraints when users input invalid or ambiguous queries, resulting in inaccurate or irrelevant recommendations.

Proposed Solution:

• Implement informative instructions, constraints, and validation mechanisms that guide users in entering valid and specific queries to obtain accurate fashion recommendations.

6. Recognition rather than recall

The Fashion GPT feature should minimize the need for users to recall information from memory by providing visible and easily accessible cues and aids.

Evaluation Findings:

• The Fashion GPT feature may lack clear labels, tooltips, or visual cues that help users understand the purpose and functionality of certain elements or controls, relying heavily on user recall.

Proposed Solution:

• Incorporate clear labels, tooltips, and visual cues within the Fashion GPT feature to aid user recognition and understanding of elements and controls, reducing the cognitive load and reliance on recall.

7. Flexibility and efficiency of use

The Fashion GPT feature should cater to both novice and expert users, offering shortcuts, customization options, and efficient interaction methods.

Evaluation Findings:

• The Fashion GPT feature may not provide efficient interaction methods or shortcuts for users to quickly obtain desired fashion recommendations or refine their searches.
• Customization options to personalize the Fashion GPT model’s outputs based on users’ preferences may be limited.

Proposed Solution:

• Implement efficient and user-friendly interaction methods, such as intuitive search functionalities or filters, to help users obtain desired recommendations efficiently.
• Provide customization options that allow users to personalize the Fashion GPT model’s outputs based on their preferences, style, or specific fashion requirements.

8. Aesthetic and minimalist design

The Fashion GPT feature should have an aesthetically pleasing design, avoiding clutter and unnecessary complexity while showcasing the generated fashion recommendations.

Evaluation Findings:

• The visual presentation of fashion recommendations within the Fashion GPT feature may be overwhelming, with information overload and distractions.

Proposed Solution:

• Optimize the visual design of the Fashion GPT feature, focusing on clean and minimalist layouts that highlight relevant fashion recommendations while maintaining an aesthetically pleasing interface.

9. Help users recognize, diagnose, and recover from errors

The Fashion GPT feature should provide relevant help resources and documentation that are easily accessible, assisting users in understanding and effectively using the feature.

Evaluation Findings:

• The Fashion GPT feature may lack accessible help resources or contextual documentation to guide users on how to effectively use the Fashion GPT model and understand its limitations.

Proposed Solution:

• Incorporate easily accessible help resources, such as tooltips, in-app guides, or a searchable knowledge base, within the Fashion GPT feature to provide users with assistance and information about its functionalities.

10. Help and documentation

The Fashion GPT feature should provide informative feedback and error messages, guiding users towards corrective actions or suggesting alternative approaches.

Evaluation Findings:

• The Fashion GPT feature may provide vague or unhelpful feedback messages when the model fails to generate suitable recommendations, leaving users unsure of how to proceed or improve their queries.

Proposed Solution:

• Enhance the feedback and error messages within the Fashion GPT feature, providing clear and actionable guidance when the model fails to generate relevant recommendations, assisting users in improving their queries or exploring alternative approaches.

Conclusion

Through the heuristic evaluation based on Nielsen’s 10 general principles for interaction design, several usability issues were identified within the Fashion GPT feature on the Myntra app.

The proposed solutions aim to address these issues and enhance the user experience by incorporating loading indicators, improving system feedback, aligning with user expectations, offering customization options, ensuring consistency, preventing errors, aiding recognition, improving efficiency, optimizing aesthetics, providing help resources, and enhancing feedback messages.

Implementing these improvements will help Myntra users enjoy a more intuitive and seamless experience when using the Fashion GPT feature, fostering engagement and satisfaction within the app.