Walmart "Imperium"

User Experience  /  Enterprise Technology

The Situation

I had the opportunity to join Walmart U.S. Omnitech to support an internal application called "Imperium."  The web application allows different corporate domains to manage business logic related to "Items and Inventory" on and their U.S., Canada, and Mexico stores.

Historically this application has been entirely developer-driven, and it is pretty clear that UX is not a priority based on its second placement in the production pipeline.  Documentation in Confluence suggests a designer came in, but the production work was lost.  Two years passed between then and my arrival, allowing for the potentiality of solving the wrong problems and overall rework.

My Contributions

I conducted qualitative research in the form of 12 off-site interviews. Then documented them, creating a sharable Mural board. Worked on improving the live application based on business stakeholder and user feedback, researched and designed a Notification Feature set, consulted on implementing user education and adoption, provided heuristic analysis, and updated UI to adopt the new design system well as establish the extension. The design system was considered a lower priority.



The onboarding process was easy because I had the help of the product owner to get me started. New users typically reach out to him to get acclimated. But that's part of the point he's often running demos that are repetitive in nature. A self-service tutorial aspect would be a great thing to introduce to the tool. I suggested delivering a tour guide when a user visits for the first time. They could also have a set of popular completed or requested items that the system could guide the user through, opening the product owner's time to more critical value-added tasks.

Onboarding generally took about 4-5 months before I felt I had a good handle on the tool and what it did. The 200+ domains involved can be a bit technical, and there's much knowledge related to each one to intelligently ask questions the right way to get the most value.


User interviews

As part of the discovery phase, I completed 12 one-on-one interviews remotely (captured in Zoom) since teams are spread all over the U.S. I had them share video so I could see body language along with screen sharing so I could associate their actions with any body language and verbal cues they may show.

See interview guide

After the interview I extracted the data and began synthesis displaying my questions along side each participant's response so I could reference them side by side within Mural.

A sanitized version of the Mural displaying interview participants. Who I sent invites to were recorded. Questions to be asked were documented and the interview results were displayed in a matriz like fashion showing the answers side by side for each person.

Improvements to live application

In order to keep things moving, development was working to apply hot fixes to the existing application as feedback was received from the business. They needed UX feedback in real time which isn't always ideal because UX needs time to investigate and research in some cases. I provided feedback in the form of annotated screens.

Fullfillment Path Flattening

One of the updates the user base was first excited about when I started was the simplification of selecting an items fulfillment path. In the current state, it was a confusing set of nested checkbox components that wasn't set up to match the user mental model as they thought through the problem.

Notification Feature Addition

Notifications were new to the system. As a brand new feature set, I reached out across organizational stovepiping to get a sense of what other products teams had already tackled in relation to their exploration and delivery. Then having that in the back of my mind, I set out to try and help streamline and align the user experience.

See Figma Mockup

User education and adoption

Users can be provided a tour similar to other SaaS products when they enter the application. I would have provided a prototype, but I didn't have time due to resources. The team wanted suggestions but only gave me time to explain verbally in a meeting which was the norm when engaging UX.

The thing I learned here is that development still relied on their intuition for how things should be. But, it still missed the mark in my opinion. They need to be provided a finished example to help place them on solid footing. What they implemented was functional, but still lacked an established look and feel to be propagated throughout entire product and provide a professional user experience complete with a firm finish.


After interviewing some of the users I began reestablishing a Personas file based on the feedback received from the sessions.
See Figma File

Heuristic Analysis

I took screen grabs of the existing applications and made annotations on top of them. It is incomplete but this was the WIP. Additionally, the IDC screens being shown are still developer driven and not user-centered driven. They are shown to give an awareness to the direction in which we were headed but not the final product.

Work In Progress Analysis

Design System Adoption (Lower Priority)

The visual look and feel was of low priority because improving the UX had a higher ROI. There were no components to use in the beginning though making this less than ideal. Everything went to a wireframe fidelity in some places because of it. However, I did get the Dashboard and Domains started.

See the Dashboard Figma

See the Domains Figma


Write up in progress