MyGNFR + Shop GNFR

Walmart associates rely on operational supplies and fixtures to keep stores, clubs, and supply chain locations running. The legacy experience made those items difficult to find, introduced uncertainty after an order was submitted, and exposed users to procurement complexity they shouldn't have needed to understand.

Weak categorization, poor search, fragmented order information, duplicate purchases, and limited visibility created workarounds that extended beyond the application. In a 24-participant survey, 22 participants said their stores borrowed items from other locations at least occasionally, and nearly half reported doing so monthly or more.

• Simple enough that a first-time requester could find the right item without procurement expertise
• Strict enough to hold up against budget rules, location permissions, and approval thresholds
• Flexible enough to serve different roles, locations, and business units without one rigid workflow
• Fast enough on the store floor through Shop GNFR, not just the fuller MyGNFR web experience
• Structured so the platform could absorb a growing roadmap, search, tracking, subscriptions, without turning into disconnected features

I became the primary UX owner for MyGNFR in June 2023 and remained the design point of continuity as the product moved through legacy replacement, enterprise rollout, and channel expansion. Early work focused on the core experience: how associates searched for operational items, created orders, and moved through approvals and tracking.

MyGNFR was one of four platforms in the indirect-procurement portfolio I owned end to end during this period. The other three were ReimburseMe, iCatalog (supplier catalog and listing), and GNFR Invoicing (supplier billing), together spanning the full list-to-order-to-invoice loop. That broader ownership changed how I approached MyGNFR. I did not treat it as an isolated tool. I treated it as one piece of a system that needed to stay coherent with the rest of the portfolio.

A single order could involve catalog and taxonomy data, item eligibility and availability, location and role permissions, budget information, approval thresholds, purchase-order creation, supplier fulfillment, delivery tracking, and exceptions. The interface was only the visible layer of a larger procurement system. A polished catalog alone would not solve unclear approvals, inconsistent item data, or missing status information.

The roadmap included search, approvals, order tracking, subscriptions, mobile access, budget information, and catalog improvements. Each request could have been designed as an isolated flow, but users experienced them as one procurement journey. I organized the product around six connected tasks: discover the correct item, evaluate whether it fits the operational need, order within location and budget rules, approve when policy requires it, track what happened after submission, and replenish recurring supplies without rebuilding the same order.

That model made it easier to see where one design decision affected another. Search quality influenced item confidence, item confidence influenced incorrect orders, order state influenced whether someone reordered, and approval visibility influenced how quickly work moved forward.

1. 01

   Separate workflow depth from frontline access

   MyGNFR needed to support detailed procurement workflows, including approvals, budget information, tracking, and exceptions, while store associates needed a faster way to order supported items on the floor. Moving the full web platform into MyWalmart would have introduced unnecessary complexity into a high-frequency store context. We maintained MyGNFR as the responsive platform for broader, more complex workflows, while extending a focused set of capabilities into MyWalmart through Shop GNFR. We accepted that the two channels would not have complete feature parity, in exchange for a clearer rule to guide future scope decisions.

2. 02

   Treat search and catalog quality as one experience problem

   The legacy system lacked effective categorization and produced poor search results, and improving the search field alone wouldn't solve incorrect or duplicate orders. I framed discovery as a combined interaction, taxonomy, and data-quality problem, pairing search and filtering improvements with clearer item information and explicit catalog requirements, because users do not really distinguish between a poor interface and poor source data.

3. 03

   Make status, approval, and next action part of the core journey

   Submission was not the end of the user journey. A request could remain editable, require approval, be rejected, create a purchase order, move to supplier fulfillment, or run into an exception. I treated status visibility and next-action clarity as core interaction requirements across carts, approvals, orders, and tracking, translating internal process state into user language without exposing unnecessary backend detail.

4. 04

   Use consistent patterns without forcing identical operating models

   The platform supported different roles, locations, business units, item types, and device contexts, and a single rigid workflow wouldn't fit every operational rule. I used common structures for navigation, item discovery, carts, status, approvals, and tracking, then let the product adapt through permissions, content, and workflow rules. This gave the platform a common language while still allowing operational variation.

An early discovery survey of 24 participants and two rounds of moderated prototype testing with 58 participants across Walmart and Sam's Club (47 in Round 1, 11 in Round 2 after revisions) validated the direction: 94% user satisfaction. After launch, ordering process time dropped from about 45 minutes in early research to about 30 seconds in the shipped product.

MyGNFR web launched in October 2024 and is now live in 100% of Walmart stores. Shop GNFR launched inside MyWalmart shortly after. Accessibility was treated as a delivery requirement rather than a final review. I coordinated with accessibility and engineering partners so issues involving labels, semantics, focus behavior, keyboard access, and error handling were addressed before release.

Owning the connective tissue

My leadership on the project came from owning the connective tissue between features, teams, and delivery phases. I reduced ambiguity when requirements arrived as isolated capabilities, built cross-functional alignment with product, engineering, procurement, catalog operations, and accessibility partners, and kept design continuity across a multi-year roadmap, so the platform evolved as one product instead of a series of disconnected releases.

Shop GNFR and MyGNFR changed how I think about where design work actually happens. The hardest problems here were never in Figma. Search felt broken because the underlying catalog data was inconsistent, approvals felt unclear because status lived across three different systems, and no amount of interface polish could fix either on its own.

Being useful meant pushing on data quality, taxonomy, and process alongside engineering, procurement, and catalog operations, not just the screens they fed into. That is the version of design leadership I want to keep practicing: influence that runs through people and systems I do not own, not only through what ships in the product.