Amazon
As part of a 24-hour Industry Day project, this initiative focused on improving the quality of customer reviews while safeguarding privacy preferences. By implementing user-friendly feedback collection methods, we aimed to deliver more meaningful insights to businesses without compromising customer trust. Our proposed solution enhances the accuracy and relevance of reviews, ultimately supporting better decision-making for both customers and companies. Through careful consideration of privacy concerns and thoughtful design, we developed solutions that ensure authentic and valuable feedback.
View Prototype
Case Study
Role
UX ( Researcher, Wireframes, Prototyping)
Client
Amazon - Industry Day Project
Timeline
24 Hours
Tools
Figma
Overview
Working alongside diverse disciplines such as Software Engineering, Data Science, and Cybersecurity, we tackled the challenge:
How might we improve the quality of Amazon reviews while respecting users' privacy preferences?
Problem Space
The current customer review system faces significant challenges, including issues with authenticity, inconsistent review quality, spam, and manipulation.
These problems undermine trust in reviews and negatively impact user decision-making, highlighting the need for solutions that ensure reliable and credible feedback.
A Closer Look
With 95% of users relying on reviews to guide their purchasing decisions, customer feedback has become a crucial factor in evaluating how well a product meets their needs.
Reviews offer valuable insights into the product's quality, functionality, and overall satisfaction, helping users make more informed choices
Improving the reliability of Amazon reviews will enhance the user experience and positively impact purchasing decisions, ultimately driving increased revenue.
Michael
Cautious Consumer
Age: 32
Nationality: Canadian
Occupation: Marketing Specialist
Location: Toronto, Canada
Education Level: Bachelor’s
"I want to trust that the reviews I read are genuine and balanced, without having to share more personal information than necessary. Convenience is key for me, but not at the expense of my privacy."
PERSONA
Goals
• To have a trusted source of reviews that accurately reflect product quality.
• To maintain his anonymity and personal privacy while contributing to the review system.
Behaviours
He wants to feel confident that the reviews he reads are authentic, balanced, and relevant.
He doesn’t want to disclose unnecessary personal details when submitting reviews or shopping.
Frustrations
He often encounters reviews that seem fake or biased, making it harder to trust the ratings.
Compromised Privacy: Michael feels uncomfortable when platforms request too much personal information or track his behaviour after purchases.
From our secondary research, we created a proto-persona named Michael. He’s an avid Amazon user and a cautious consumer who loves sharing his experiences with products and helping others through his reviews.
Approach
To address the problem of improving the quality of Amazon reviews while maintaining user privacy, several strategies can be employed based on secondary research. We focussed on enhancing review credibility through machine learning. By applying text analysis, frequently used keywords and sentiments can be identified to filter out low-quality or misleading reviews. This can also help flag suspicious or spam reviews. Our Data Scientists proposed to apply Recurrent Neural Networks (RNNs) to determine the importance of specific words, enhancing the accuracy and quality of sentiment analysis in reviews.
Generate review highlights by extracting key terms, providing deeper insights into the overall rating based on these keywords.
Review Writing Prompts
Offering review writing assistance simplifies the process for customers, making it easier for them to provide feedback while improving the quality of their reviews. This can involve tools like prompts or templates that guide users through structuring their thoughts more clearly. Such support encourages more participation and ensures reviews are detailed and helpful, reducing incomplete or vague feedback. As a result, this enhances the overall reliability and usefulness of reviews, benefiting both shoppers and businesses by promoting better purchasing decisions.
Additional Review Filters
We proposed to introduce additional filters that will allow users to sort and view reviews based on specific keywords. This feature will make it easier for the users to find relevant feedback by highlighting the most important terms users are searching for, helping to streamline the decision-making process. By focusing on keywords, it becomes much more convenient to filter through large amounts of reviews and quickly find the insights that matter most.
Rating Evaluation
By incorporating data-driven rating evaluations, we can significantly improve the trustworthiness and credibility of both customer reviews and Amazon ratings. This approach uses data analysis to assess reviews more objectively, reducing the impact of biased or manipulated feedback. As a result, users will have greater confidence in the ratings they see, knowing that the information is backed by a more reliable and transparent system. This boost in authenticity will not only help customers make more informed purchasing decisions but also enhance Amazon's reputation as a trustworthy platform for product reviews.
Verified Account
Mark of Distinction.
Frequent reviewers who garner a high number of likes will receive a special recognition called the "Mark of Distinction," which certifies their reviews as authentic and trustworthy. This initiative encourages more meaningful feedback and helps users identify reliable sources of information, ultimately enhancing the credibility and quality of the review system on the platform.
Key Takeaways
Research and Ideation
During the early stages of our project, collaborating with my group multiple times ensured we were all aligned on our overall objective. This early alignment was crucial in helping us form smaller teams, where me and my UX designer partner, worked closely with both front-end and back-end developers. This collaboration ensured that our features were in sync with our design and functionality goals. Investing that initial time in brainstorming and ideation really paid off, resulting in a more streamlined production process.
Designing with Empathy
I started the process by adopting the perspective of the customer. Together with my UX partner, we focused on fully understanding the user journey, ensuring that users were central to our approach. This allowed us to explore the problem and solutions from both detailed and broader viewpoints. By prioritizing the user experience, we enhanced not only the functionality of our solutions but also how they emotionally resonate with customers. This dual approach helped us pinpoint key touchpoints and pain points, resulting in more thoughtful design decisions that address actual user needs.