.png){kind=icon}
.jpg)
CDF Amazon Project
Description
In my role as Senior UX/UI Designer, I had the opportunity to lead a pivotal project at Amazon, aimed at improving the Customer Delivery Feedback (CDF) system. The primary objective was to enhance the user experience by redesigning existing features and introducing self-resolution links, ultimately reducing customer service overload and empowering users to address their delivery issues independently.
​
Project Overview
Based on the last research and analysis customer service was overwhelmed with a high volume of incoming requests related to delivery issues. This overload was leading to delays in response times and frustrated customers.
The absence of self-help resources or guidance within the CDF system meant that users had to contact customer service every time they faced these recurring problems.
​
Project Goals
In response to these challenges, our primary project goals were:
-
Reduce Customer Service Overload: Implement solutions that would alleviate the customer service team's burden and improve their ability to handle complex issues promptly.
-
Empower Users with Self-Resolution: Introduce self-resolution links and resources within the CDF system to enable users to address and fix common delivery problems independently.
-
Enhance User Experience: Redesign the CDF system to optimize the overall user experience, making it more user-friendly, intuitive, and efficient in guiding users toward resolutions.​
By addressing these goals, we aimed to create a win-win situation: alleviating customer service overload while providing users with a more satisfying and efficient experience when dealing with delivery-related issues.
​
Customer's goals
-
I want to resolve the issue quickly
-
I want Amazon to fix the issue to prevent it from happening again
-
I want to provide Amazon with the details about the issue I faced
-
I want to know that Amazon is working on / has fixed my issue
​
Amazon goals
-
I want to understand the details of the issue the customer faced
-
I want to ensure we have a systematic way of tracking and fixing the issues reported by the customer
-
I want to confirm with the customer that their issue was resolved
Ingress points
The Customer Delivery Feedback (CDF) flow begins when a customer clicks on thumbs up or thumbs down from any of these 4 ingress points: the Progress Tracker page (PT), the Product Owner page (POP), as well as Delivered email and push notifications. In this project, we are improving the CX of the negative delivery feedback workflow i.e., when customer clicks on thumbs down.
Current flow
Today, customers can provide positive or negative feedback by selecting thumbs up/down (on L1 page). Once a customer selects a positive or negative response, they can provide details through a set of pre-defined responses (L2 reason codes). After the customer selects an applicable reason code and clicks on submit, the customer is presented with options to self-resolve their issue (e.g., return an item, update delivery instructions, or contact customer service).
User Scenarios
Proposed Happy Path
We propose giving customers an option to provide additional details after they have provided initial negative feedback (L2 reason codes) through an additional set of pre-defined reason codes (L3 reason codes).
Tree test
The goal of this study was to understand how users interact with the negative delivery feedback process and how they locate the information they need based on the scenario. Leaving negative delivery feedback requires customers to correctly categorize their issues by following a series of prompts. See the flow below for an example.
We asked 21 participants to complete a tree test we gave them an example of a potential negative delivery experience. For example, one task asked the participants to locate where to find the input to the issue: “Package was delivered to a neighboring address.” The participant then had to select two levels down within the “tree” to find the correct location. If the participants found the correct location, then the IA is said to have good findability.
.png)
Based on the collected data from contributors and their click interactions, we can determine that the average number of clicks per session is 34. It means that participants encountered challenges when attempting to locate specific content within the tasks (initially we have 8 tasks and 8 solutions):
Main Areas of Improvements
-
Content organization: Participants had difficulty locating specific content. Recommend: reorganizing the content structure.
-
Content relevance: Participants struggled to find relevant content. Recommend: revising or removing the irrelevant content.
-
Content clarity: Participants noted content that was difficult to understand or follow. Recommend: revising the content to make it clearer and more concise.
Unmoderated Usability Testing
After the design phase, I conducted Unmoderated Usability Testing (both static and dynamic) to gain insights and understand pain points. ​
​
Goals:
-
Any confusion or difficulty they have understanding elements on the page.
-
Whether they find the options clear and easy to understand.
-
If they believe that providing feedback can help improve the delivery experience.
-
Any suggestions they have for adding, removing, or changing things on the pages.
-
If they understand that the PoM is interactive.
-
If they understand how to get out of the flow if they don’t wish to proceed.
-
If they understand the follow-up questions and why they’re getting them.
​
Target Audience:
-
Recent recipients of orders from Amazon.
-
Familiar with Amazon delivery experience.
-
Located in the United States.
-
Provided delivery feedback before.
​
By conducting the usability testing for the “Pin on Map” project, the goal was to gain a better understanding of how users interact with the process and whether they could easily navigate through the process, and make corrections to avoid similar problems in the future. ​
We wanted to identify if users faced any difficulties and if they understood map interactivity, where they can click to make pin corrections.​
The testing was unmoderated, users received verbal and written instructions as well as multiple-choice questions.
We asked them to imagine a scenario where their delivery was mistakenly delivered to a neighboring address and they should provide their thoughts and steps on each task.
This testing method allows us to evaluate user interactions, gain insights, and identify any problems and areas of confusion or potential improvements.
​
The primary objectives were as follows:​
Regarding L2 + L3:
-
Test a scenario where users select two L2 reason codes and progress to consecutive L3 pages.
-
Assess users' comprehension of the follow-up questions and the reasons behind their presentation.
-
Evaluate whether users understand how to exit the flow if they choose not to proceed.​
Regarding "Pin on Map": The central focus of this testing was to evaluate the self-resolution page and ensure users' comprehension of the content. Here are some example prompts:
-
On the L2 page, ask users to describe what they see.
-
Inquire about their expectations if they were to click on the map.
-
Seek their understanding of the other fields on this page and what they represent.
-
Assess how they would proceed if they did not wish to make any changes.
-
Gauge their emotional response when encountering additional L3/PoM/Thank you pages.
-
Identify any confusion or difficulty they experience in understanding elements on the page.
-
Evaluate whether users find the options clear and easy to understand.
-
Explore whether users believe that providing feedback can contribute to improving the delivery experience.
-
Encourage users to offer suggestions for adding, removing, or modifying elements on the pages.
​
By conducting this comprehensive usability testing, we aimed to gather valuable insights and ensure the effectiveness of the design in guiding users through the process, making it user-friendly, and providing opportunities for self-resolution while addressing any potential pain points or confusion.
​​​
Top-Level Recommendations
Based on the usability testing, we have identified key areas for improvement:
More than 94% of participants found the "PoM" process clear and easy to understand. They correctly selected the "Change your location on the map" option to resolve their issue.
However, it was observed that participants didn't recognize that the map was interactive and clickable.
We can consider two options: keeping the current version or adding a title "Click on the map to change your location" as a title for the map.
​​​
Enhancements for a More Engaging and Intuitive User Experience​​
Streamlined Navigation: Recognizing that users found the lengthy process tedious, I reimagined the navigation to make it more efficient and enjoyable.
Simplified Categorization: To address the issue of confusion caused by numerous categories and sub-categories, I reorganized the content into a more intuitive structure. ​​
One-Click Buttons: I introduced one-click buttons strategically to expedite the process. Users can now achieve their goals with minimal clicks, reducing frustration and boredom.
Visual Division of Options: To enhance clarity and user engagement, I visually divided the primary categories, making it easier for users to identify and select.
Visual Appeal: The redesign focused on creating a visually appealing interface that not only engages users but also maintains brand aesthetics and consistency.
Results
The findings from this case study underscore the critical role of content organization, relevance, and clarity in enhancing the user experience. By implementing recommended strategies such as revising or removing irrelevant information and improving content clarity, the usability and effectiveness of the material were significantly enhanced, resulting in a more positive and satisfactory user experience.
​
The improvements made to the user experience were primarily driven by user feedback and aimed to:​
-
Decrease User Boredom: By minimizing the number of clicks required to complete tasks, the design was made more efficient and enjoyable for users.
-
Mitigate User Confusion: The reorganization of categories and sub-categories resulted in a more straightforward and intuitive navigation process.
-
Enhance Efficiency: The introduction of one-click buttons streamlined user interactions, simplifying tasks and reducing frustration.
-
Improve Visual Aesthetics: The visually appealing interface not only captivated users but also reinforced the brand's image, contributing to a more engaging user experience.
​​​
In conclusion, these results demonstrate the effectiveness of user-centered design approaches in addressing usability challenges and enhancing the overall user experience. By listening to user feedback and iteratively improving the design, organizations can create interfaces that are more efficient, intuitive, and visually appealing, ultimately leading to greater user satisfaction and success in achieving their goals.
Thank you for your attention!