During my internship at Spreetail, I worked as a UX Designer in the Listing Health team. Spreetail is an e-commerce business that lists thousands of items online through various channels every day. The team develops internal software to help manage the item listings on the Amazon Marketplace.
Every day, the Listing Health Tool detects thousands of inactive listing that is taken down from Amazon Marketplace due to various reasons. The Listing Health Management team is tasked with addressing these inactive listings to minimize listing downtime and loss of revenue.
The goal was to design a dashboard that helps the listing health team quickly identify high priority inactive listings and take appropriate action and see the overall picture of all the inactive listing on the Amazon Marketplace.
I conducted user interviews and card-sorting activities to understand the problem context, used the insights found to ideate and generate low fidelity concepts to quickly gather feedback. After iterations based on user feedback, I refined the concepts and applied the company's design systems to deliver a production ready mock up and interaction flow. The design was later implemented by the software team.
4 weeks
Overwhelming amount of data comes in as thousands of rows in the table of the Listing Health Tool. A lot of time is spent compiling data from different sources into reports.
Managers don't have the time to dive into each of the product categories and monitor all of the data. It's very difficult to single out high priority issues that needs to be dealt with.
Users can't really act on the data in real time without relying on each of the specialist filtering the data.
Through user research with card sorting and interviews, I was able to understand the user's workflow and decision-making process in identifying and addressing inactive listings.
I was then able to generate concepts from these insights and iterate on the design based on user feedback.
The dashboard highlights important metrics that can be utilized to quickly identify key inactive product listings. The dashboard also provides the users with monitoring tools to see changes that occur, allowing users to quickly spot anomalies in product listing behaviors. Statistics for the sub-team within the Listing Health team are also clearly compartmentalized to allow managers to quickly assess the performance of each sub-team as well.
The was dashboard planned to be implemented in the Listing Health Tool in the following sprint.
From this project, I learned how to communicate with and manage requirements from different stakeholders and what I needed to consider when handing off my designs to developers.
Every day, the Listing Health Tool detects thousands of inactive listing that is taken down from Amazon Marketplace due to various reasons. In order for the business to address these inactive listings, the Listing Health Management team is tasked with addressing these inactive listings to minimize listing downtime and loss of revenue.
I started the design process for the dashboard by diving into the Listing Health Tool and exploring it thoroughly as well as reaching out directly to stakeholders and users of the tool. After initial conversations with the Listing Health team, I found that there were multiple users that interacts with the platform.
I realized that I needed to learn more about not only how the tool is being used, but also how the users are making decisions through the tool as well. I conducted several virtual contextual interviews with the users group to understand their workflow and decision making.
I also set up a quick card-sorting activity with different types of data as the cards and let the users categorized the types into different “buckets” to understand the relationship between each type.
The users can be divided into two groups based on their role in the Listing Health team: the Listing Health Specialists, and Managers.
The Listing Health Specialists responsible for spotting, tracking, and fixing inactive listings on the Amazon Marketplace. Each specialist is responsible for a group of inactive listing reasons, usually 3 to 5 per person, and will only address inactive listings due to the reasons they're responsible for.
Within the team Listing Health team, one or two Managers oversee the operation of all the inactive listing the whole team is responsible for.
Listing Health Specialists
Overwhelming amount of data comes in as thousands of rows in the table of the Listing Health Tool.
Even if they sort or filter the data, they still need to take a really long time to go through the individual issues. They only see the high priority issues when they filter the data manually.
Listing Health Managers
A lot of time is spent compiling data from different sources into reports. They don't have the time to dive into each of the product categories and monitor all of the data.
They can't really act on the data in real time without relying on each of the specialist filtering the data.
With this information, I realized that I needed to learn more about not only how the tool is being used, but also how the users are making decisions through the tool as well. Therefore, I conducted several virtual contextual interviews with both the users group to understand their workflow and decision making.
During the contextual interviews the participants were asked to demonstrate how they interact with the Listing Health Tool in order to address the inactive listings that they're responsible for. Then follow-up questions were asked to probe into how the participants process the information available on the tool.
From the interviews, I quickly realized how the perspective of the Managers differ from the working Specialists. I also understood more about the context of how the users use the tool, what information do they care about, what tabs to they have opened while they're working, and how people prioritize listings that need to be addressed.
Specialists...
Managers...
Therefore, from the findings, I decided to focus on the Managers as the primary user group for the dashboard.
I decided to focus on the Managers as the primary user group for the dashboard since they care more about the overall pictures and needs to quickly identify key high-priority listings for the team to work on.
I started the ideation process by brainstorming and sketching my ideas. Then I narrowed down the sketches that I felt had potential and identified how the ideas answers the users' needs. With some of the ideas narrowed down, I asked for some user and designer feedbacks for the ideas.
With ideas and some feedbacks from stakeholders and other designers on the sketches, I began making prototypes in Figma, which were used to further validate the design.
The goals for the prototyping process were to...
The final prototype were high-fidelity and built with the official design system used for the tool. This provided the benefit of clearer communication to the stakeholders, allowing for concise feedback with regard to the user's expectation of the look and feel of the tool in the working environment.
The design of the dashboard highlights important metrics that can be utilized to quickly identify key inactive product listings. I specifically designed the dashboard to emphasize the inactive listings that has the biggest impact on revenue. The dashboard also provides the users with monitoring tools to see changes that occur, allowing users to quickly spot anomalies in product listing behaviors.
Statistics for the sub-team within the Listing Health team are also clearly compartmentalized to allow managers to quickly assess the performance of each sub-team as well.
To make sure that the design is ready for implementation, the design is thoroughly crosschecked with the company's design system. Every developer in the team involve with implementing the design were also briefed regarding the user flow and interactions, which were also illustrated on the design file in Figma.
I made sure to involve all the relevant stakeholders and developers in the design presentation and delivery so that both the developers and users know what to expect from the design.
As I'm learning to work with stakeholders and developers in a software team, I'd like to share some of my learnings from this experience as well. For starters, I learned the importance of good communication within the team. This is not only for communicating with the stakeholders about their wants and needs, but also communicating with developers and understanding how the tool is built. In this case, it is making sure that I understood where the data is coming from, how the elements behave, and how the web-app was built.
I also learned that for prototyping my design, using real and familiar numbers in the prototype can help users understand what they're seeing in the design. To elaborate, in a highly specialized use-case of a tool such as the Listing Health Dashboard, users will infer what the design is saying based on the numbers that they're used to seeing. For example, if they're used to seeing the number of inactive listing for the Compliance category being in the range of 200-500 issues, then when they see a number of, for example 384, they will quickly assume the significance of that number. This is definitely something to keep in mind when designing with in the context similar to this, where the users are highly skilled specialists who is very familiar with their job.
