Fusion Pricing: Retails Search � Tracy Lindstrom

6.4 / 7

Usability score post-redesign

6.2 / 7

Satisfaction score � Product Requirements Document rewritten before dev began

Features shipped � research influenced cross-product navigation

The problem

Pricing managers at Sam's Club couldn't act until they'd done research the tool should have done for them. Finding the right items meant leaving the platform entirely: opening Excel or SQL, building a working list, then returning to price.

Once inside the tool, filters reset every session, approval context was missing, and results weren't ordered by relevance. Ninety-five percent of price changes were still flowing through overnight spreadsheet templates on a platform designed to replace them. On a system processing 83 million price changes a year, that's not a UX problem. That's a business problem.

"Every power user on the team was blocked by the 20-item cap. Nobody could search by Sign ID. Filters reset every session, so everyone started over every time they logged in."

Composite finding from five moderated usability sessions, 60 minutes each.

What changed

Four concept directions all failed at exactly the same point. That wasn't a design problem � it was a signal that the requirements hadn't accounted for how pricing managers actually worked. The session recordings went directly to the product team. The Product Requirements Document (PRD) was rewritten before development began. What shipped reflected the real workflow, not the assumed one. The research findings also traveled beyond the original project, reshaping global navigation across the broader Fusion ecosystem. This was the first formal usability study the Pricing org had ever run.

Research & discovery

Four compliant concepts, brought to usability sessions to find the ceiling of what the existing requirements could support. Every concept failed at the same point � not because the designs were wrong, but because the requirements hadn't accounted for how pricing managers actually worked. Session recordings went directly to the product team. The Product Requirements Document was rewritten before development began.

Key findings

The exit-to-Excel pattern was load-bearing. Users weren't leaving the platform out of habit. They were leaving because it couldn't do what they needed. The workaround was structural.
Filters were ephemeral. Every session started from scratch. For pricing managers working large item sets across multiple sessions, this created significant overhead.
Search lacked hierarchy awareness. Pricing at Sam's Club is hierarchical: you price a category, a subcategory, a product line. The search returned flat results that didn't reflect how the work was actually organized.
Requirements had a blind spot. The existing PRD was written without input from observed working sessions. Four distinct design directions all broke at the same point, which was the signal that we had a requirements problem, not a design problem.

Retails search panel with item number query showing 12 flat results including Folgers coffee, Member's Mark paper towels, bath tissue, and Christmas tree items with checkboxes — Item-number search returning a flat list � the starting point. Pricing managers used item numbers as a primary lookup method, but results came back as an undifferentiated list with no hierarchical context. No way to price a base variant and catch all its sub-items at once.

Design process

The first formal usability study the Pricing org had ever conducted. Running four concepts in parallel was deliberate: the goal was surfacing requirements gaps, not just design preferences.

Key design decisions

Persistent filters. Filter state now persists across sessions. Pricing managers working large item sets no longer rebuild their context from scratch on every login.
Sign ID search. Added direct search by Sign ID, a lookup pattern pricing managers used constantly outside the tool and couldn't do inside it.
Hierarchy-based navigation. Search results are organized to reflect the actual structure of pricing work: category, subcategory, product line. Not flat item lists.
Bulk selection patterns. Built for real working scale: pricing managers often need to act on dozens of items at once, not one at a time.

Retails search panel with a product code search showing 38 results organized hierarchically by base variant � Member's Mark men's essential crew neck tee � with collapsible product groups for Black small, medium, large, and x-large variants, each with item-level checkboxes — Hierarchy-aware search results � the core design solution. Results are organized by base variant, then product, then item, matching how pricing managers actually think about their work. Select a product node and all items in the hierarchy are captured. No more pricing each size individually.

Retails search panel showing a Protein attribute search with Member's Mark protein shake hierarchy expanded and checked � 4 items selected counter in header, both chocolate and vanilla variants highlighted with blue checkboxes, and 20 items selected shown in the main pricing table behind the panel — Bulk selection in action � the "Protein" attribute search surfaces all protein-category items, and checking the hierarchy node captures all variants at once. The 20-item counter in the main table updates in real time. This is the workflow that replaced the overnight spreadsheet template.

Outcomes & impact

6.4/7 usability, 6.2/7 satisfaction. Four features shipped. Research findings expanded beyond scope, informing global navigation across the Fusion ecosystem. The process became a model � not just the output.

6.4 / 7 usability score
6.2 / 7 satisfaction score
PRD rewritten before development began
4 features shipped
Research informed cross-product navigation improvements across Fusion
First formal usability study in the Pricing org
9.25 / 10 peer recommendation score