Optimizing Inventory Replenishment: Building an AI-Assisted Decision Engine
Transforming complex inventory replenishment workflows from a passive administrative tool into a high-impact AI decision platform.
Role: Lead Product Designer (Contract)
Scope: Product Framing, UX Audit, Information Architecture, Interactive Flow & High-Fidelity UI
Timeline: 3 months
Problem Framing & Context
Core product challenge: Transforming a passive, data-dense inventory table into an active AI decision engine for supply chain teams.
Nexora’s original replenishment experience surfaced raw inventory data but failed to translate complexity into clear, actionable decisions for enterprise supply chain teams.
The Problem: Legacy workflows prioritized operational configuration over prioritized action, creating high cognitive friction and hiding strategic business value.
Target User: Merchandising Planners and Inventory Managers responsible for balancing stock availability, preventing stockouts, and reducing overstock across retail networks.
How Might We: How might we help merchandising teams quickly identify, prioritize, and act on replenishment opportunities that maximize revenue and reduce inventory risk?
Strategic UX Audit & Shifting the Flow
A systematic audit revealed that users were forced to manage operational complexity before experiencing any differentiated product value.
To solve this, we restructured the core experience around a 3-pillar framework:
Discover: Strategic visibility and goal setting before setup.
Configure: Streamlined, goal-led parameters and scope definition.
Decide: Target-aligned action clarity driven by AI recommendations.
How might we help merchandising teams quickly identify, prioritize, and act on replenishment opportunities that maximize revenue and reduce inventory risk?
User Flow
Shifting replenishment from manual configuration to AI-supported decision-making.
High-Fidelity Solution Walkthrough
01. Strategic Dashboard (Discover)
Instead of a flat admin table, the main view surfaces high-level business risk indicators—including Revenue Opportunity, Stockout Risk, and Overstock Risk—allowing managers to filter plans by strategic urgency.
02. Goal-Led Plan Setup (Configure)
Users begin by selecting a primary business goal (e.g., Revenue Growth or Stockout Prevention) alongside scheduling parameters, aligning the AI algorithm with their immediate operational strategy.
03. Contextual Scope & Real-Time Preview (Configure)
A structured 4-step scope builder allows planners to specify primary warehouses, geographic coverage, and product categories while displaying an instant live preview of impacted sites and SKUs.
04. AI Recommendation Review & Execution (Decide)
Before submitting, the system presents projected revenue uplift, prevented stockout risks, and financial savings. High-density data tables highlight stock availability constraints and project post-replenishment balance in real time.
Outcome / Product Impact
Product impact: Transformed manual replenishment setup into an intelligent, goal-driven AI decision platform.
Reduced Cognitive Friction: Streamlined complex multi-location inventory configuration into an intuitive 3-step guided flow.
Increased Decision Confidence: Transparent AI projections provided clear financial and stockout risk metrics before execution.
Key Takeaway: High-density B2B interfaces excel when complex raw data is filtered through clear business goals, predictive AI insights, and transparent outcome projections.
Reflection
Key takeaway: High-density B2B interfaces excel when complex raw data is filtered through clear business goals, predictive AI insights, and transparent outcome projections.
This project evolved inventory replenishment from a passive administrative table into an AI-supported decision engine. Rather than simply displaying stock metrics, the work focused on restructuring how planners evaluate risk, balance multi-location inventory, and execute high-stakes orders with confidence.