Enhance the CV search experience for Shopee App
Introduction
The project’s goal is to enhance users’ experience when using Shopee Computer Vision (CV) search process. And 2 questions need to be answered to reach the goal:
What do active CV users think of the Shopee CV search?
How to improve their experience?
The Process
“In the digital age, creating meaningful design requires us to understand people who are different, anticipate what they will want to do, and provide them with the tools they need, exactly when they need them.”
— Jeffrey Zeldman
Below are the methodologies included in each phase of the design process:
Empathize (why?)
- Desk Research
- Survey & Questionnaires
- Contextual Inquiry
- Heuristic evaluation
Define
- Affinity mapping
- Personas
- Journey mappings
- Problem statements
Prototype
- Wireframe design
- Prototype design
Test
- Interactive prototype
- Visual mockup
The Research
The project conducted a Survey & Questionnaires with 39 respondents. Of the 39 only 17 respondents are aware of or have performed Computer Vision search function. The 15 respondents would form the research interest group and insights from the findings are listed below:
67% of the Users are Female.
83% uses Mobile phone for e-commerce shoppping.
Most important function to them is search result accuracy.
Contextual Inquiry
10 of the interviewees was chosen, they were given a task for them to carry out, which was screen-recorded for further analysis.
Buy an item using the CV search function on the Shopee app
- Generally, the users didn’t have any trouble locating the CV search icon on the Home page.
- 7 /10 users took a photo of an item with a cluttered background.
- 9/10 users feedback that the accuracy of the result page could be further improved. They are not able to find an exact match of their item of choice.
Personas and Journey Mapping
Based on the key findings and research on the app, 1 persona was created to depict the user’s behaviour, sentiment and pain-points throughout their stages of using the app.
Key Results and Findings:
Design Iterations
With insights from the key results and findings, the wireframe based on the proposed features was created:
Now!!! The proposed solution in action:
Thank you
I would like to take this opportunity to thank my 3 other teammates Jason, Ser Han and our talented designer Xiuwen.
