Amazon Returning Process User Research

Yuning Zhang, Isabel Karolczak

Tips: I tried to present more thinking process and discussing details in 'Question Planning', 'Prep the Survey for Launch: Survey + Pretest', and 'Summary Findings and Lessons Learned' sections by checking the discussion notes we took. Because we collaborated on the synchronized google doc, so we kept a final doc instead of every iteration version.

Introduction

Amazon.com is one of the largest e-commerce websites in the United States. However, we want to take a deeper look into consumer behavior in relation to how users return items that they have ordered. Are some users more intimidated by returning items that they do not want? Does it impact what kinds of items that they buy online? In what way?

We want to understand how these kinds of thought processes impact how users feel about and interact with Amazon's item return process.

Questionnaire Planning

co-working

Step 1: Got the blank template by reading the book 'Complete Guide to Writing Questionaires' by David F. Harris, and divided the sections into 'Screener', 'Overall Satisfaction', 'Return History', 'Customer Expectations', 'Measures of Competitors', and 'Demographics'.

Step 2: Co-worked on it.

Step 3: Critiques and discussion. Conversation examples:

"Isabel: I don't think we need both 'overall satisfaction' and 'step's satisfaction' --- I think we can determine steps's satisfaction through asking questions about their experiences or struggles with the return process so we just need to keep the 'overall satisfaction'

Me: I am not sure if we can use open-ended question in the questionnaire section, I remember the book said we'd better use multiple choices or rate sth from a scale question in questionnaire to keep it simple and involve more people. Let me check the book and internet!

Me: Oh the book did mention we could use open-ended question in the questionnaire. And I searched on the internet, it showed getting multiple potential responses from the open-ended questions can give room for further probing by the moderator. You are right, we can do that!

Isabel: Awesome!"

"Me: My previous description in data analysis part in the second screener may not make sense. For instance, users who quit Amazon choose shipment as their most dissatisfied factor versus users who has issues with return process but still shop on Amazon choose seller attitude then we may deduce shipment is the cause for users to finally drop the Amazon. I prefer to keep this one since Amazon may want to see which factor is the last straw.

Isabel: That makes sense! I'll change it back to that description."

Step 4: Finalized questionnaire planning.

Prep the Survey for Launch: Survey + Pretest

co-working

Step 1: Draft (Page 2-5 of google doc that links below)

Step 2: Critiques and discussion. My critiques examples:

  • Changed the question 'Please let us know how much you satisfied with the overall return process in the last year' to 'Please let us know how much you satisfied or dissatisfied with the overall return process in the recent 12 months.'

Reason: mention both sides (satisfied & dissatisfied) to avoid bias and use 'the recent 12 months' instead of 'the last year' for clarification since it's ambiguous when to start counting.

  • Changed the question 'On a scale from 1 to 5, how difficult were these aspects of Amazon’s return process for you? ( 1 = This process was very difficult for me, 5 = This process was very easy for me)' to 'How will you rate these aspects of Amazon’s return process for you, yourself? ( 1 = This process was very easy for me, 5 = This process was very difficult for me)'

Reason: 'How difficult' contains bias since you assume the users meet some levels of difficulty, it's better to use 'How will you rate' which doesn't have bias and preference; ask questions before describing scales to avoid confusion; we use 1 for easy and 5 for difficult for convention.

Step 3: Put the draft into google form.

Step 4: Did the pilot session and took notes to improve the survey before it launched (Page 1 of google doc that links below).

Step 5: Finalized the survey in the google form (Google form is linked below).

Prep Your Survey for Launch

Study Plan and Recruiting Screener

co-working

Study Plan and Recruiting Screener

Discussion Guide

co-working

Discussion Guide

Moderated Usability Study

Isabel Karolczak: p1-p4, me: p5-p6

Example p5:

video1283439019.mp4
P6.xlsx

Summary, Findings and Lessons Learned

Isabel Karolczak: Overall Findings 1-4 (i.e. p7-p10)

Me: Overall Findings 5-9 (i.e. p11-p18)

Others: Co-working

Purple and White Marketing Presentation.pptx

Example (Overall Finding 5): When seeking out outside help, participants struggled with identifying where to go

This was one section of our usability test, we asked users 'how would you get into contact with Amazon to ensure that your return requests has been processed'? Basically we were asking them to show us how to find the Amazon customer service, here are two video clips.

P5 - Google Drive - Google Chrome 2022-11-13 06-59-51 - Trim.mp4
p6 - Google Drive - Google Chrome 2022-11-13 07-08-30 - Trim.mp4

I found both participants scroll down to the bottom of the page to find the link of customer service but it turned out they both failed to find it here. After some struggling, they finally found the tab on the order history page. It showed users' habit to find links at the bottom of the page, but the 'help' button here is not clear and intuitive. So our recommendation is to use a more obvious word instead of 'help' at the bottom of the website page to inform users it was actually the customer service tab.

Lessons I learned from usability test:

Some participants felt stressed during the usability test, they feared to give me the impression that they were not smart enough to find the correct buttons – would try harder than normal to complete the tasks, which may affect accuracy.

We should mention in the instruction part that 'if you cannot find anything, it is not your fault, it is the UI’s fault'.

Reflection

I was very excited to do this project. As a heavy user of online shopping platforms, this project gave me a chance to take a deeper look at how users feel about and interact with Amazon's item return process, especially which steps users feel struggled with and which parts of the Amazon website confused users. We went through every phase of user research to learn how each section worked, including questionnaire planning, survey, study plan, recruiting screener, discussion guide, and moderated usability study. And we were excited to get valuable findings and lessons in the process.

I appreciated having a perfect teammate Isabel. She seldom did online shopping, but she supported me when I came out of this topic and showed great interest and responsibility in completing the project. We had a great time discussing and co-working on many sections.

One point we can improve was to get critiques and feedback in each phase more broadly and efficiently. We relied on the internet, books, and team discussion. Later we realized that directly talking to user researchers in the industry and connecting more with users are important methods.

Overall we were proud of our first user research project and we believed we could do better in the future!