Yiwei Lin UX Design Portfolio

Chatbot

E-commerce, Product Design, Responsive Web App & Native Mobile

 My Role

  • Product Design Lead: build and maintain the seller's tool from scratch. Roadmapping. Cross-team communication. Stack holder management. Directly present to VP and CIO.

  • UX Design: Site Map & User Flows, Lo-fi to Hi-fi Wireframes, Usability Testing, Dev Hand-off

  • UI Design: Responsive Design, Prototyping

Design Tools

  • Design: Figma, Photoshop, Keynotes, Sketch

  • Team collaboration: Jira & Confluence, Miro Board, Teams, Zoom

Time Duration

  • 3 months (6 Weeks for this case study)

banner_light green_chatbot.jpg

Introduction

Business Goals & Planning

 

Reduce the agent labor cost

Create conversational experience through chatbot. Reduce the frequency from the salesforce agents.

Increase customer satisfaction

Offer omni-channel support. Ask for feedback across all touch points.

Roadmapping document (2021)

 

According to the roadmapping above, I conducted the 6 weeks design planning for this projects.

 

6 Week Design Planning


Introduction

Problems and Opportunities

 

User Research

Collaborated with user researcher for the 20+ user interviews and created 3 personas.

 

Research Findings

What are significant tasks users are looking for solving from the chatbot?

  1. Orders and Returns

    (Check order status, cancel order, initiate returns, check refunds)

  2. Customer Satisfaction

    (Review process, Waiting period, Loading animation)

What’s most important information when users are looking for an order?

Items Information

Photos of items, item title (what),
Item price
Stores/purchase methods (how)

Order Information

purchase date (when),
order number - search (Unique Identifier)
return case number or item skus (Unique Identifier)

banner_light green_chatbot.jpg

Design Solution 1

Orders and Returns

 

Site map for orders & return section


According to the research finding, I transformed the original website page to 640 px chatbot screens.

 

Figma Prototype - Orders


In this flow, I have collaborated with the back-end developers to map the new order status to a more user-friendly status in the user interface.

Here are some research finding I used:

  • Customers don't understand the logistic terms.

  • Customers don't need to know the fulfillment process in warehouse, while agents can reference some information in order to provide better answers to users

  • Need to provide refine status with actions

 

Also, based on different order types and order status, I collaborated with the content writer and created different messages to customers.

 

Design Solution 2

Customer Satisfaction

How to improve customer satisfaction?

= What’s the factors will impact the customer satisfaction?


Design Decisions

  1. Ask the right questions when collecting user feedback.

  2. Differentiate the agents', designers' or developers' success metrics

 

Designer's Success Metrics

  • Task Completion - First Contact Resolution (FCR)

  • Customer Satisfaction Score

  • Average Timespent on Each Flow

Other Metrics

  • Average Call Abandonment Rate

  • Percentage of Call Blocked

  • Average Time in Queue

Agent's Success Metrics

  • Service Level

  • Average Speed of Answer

  • Average Handle Time

  • First Contact Resolutions

  • Occupancy Rates

  • Peak Hour Traffic

Figma Prototype - Customer Satisfaction


banner_light green_chatbot.jpg

 Impact

 

 

What I have learned

  1. The best practice of conversational UI

  2. Collaboration with content writers and researchers

  3. Components, Auto-layout, Design System
    Documentation
    Communication