CASE STUDY #1
Improving Product Discovery Through AI-Supported Search & Structured Data
Context
Within a complex B2B eCommerce platform, product discovery was a primary driver of customer success. Users relied heavily on search and navigation to find highly specific products across a large catalog.
The Problem
Search results were inconsistent, and navigation lacked alignment with how users actually thought about products.
- High “no results” scenarios
- Misalignment between metadata and user intent
- Over-reliance on manual content structuring
My Approach
I treated search and navigation as a system, not separate features.
The focus was aligning:
- User intent
- Structured product data
- Search engine logic (Coveo + internal models)
Rather than incremental fixes, I pushed toward a scalable, rules-driven foundation.
What I Did
- Led improvements to search relevance using AI/ML-supported capabilities
- Defined structured metadata strategy to better align with user intent
- Improved faceted navigation and filtering logic
- Partnered with engineering and data teams to refine indexing and ranking logic
- Used GA4 and CSAT signals to continuously evaluate and adjust
Outcome
- Improved product discovery experience and reduced friction in search journeys
- Increased alignment between user expectations and results
- Enabled more scalable and maintainable search and navigation systems
Why It Matters
Search is one of the highest-leverage areas in digital commerce.
This work improved both customer outcomes and internal efficiency, while creating a foundation for future AI-driven enhancements.
CASE STUDY #2
Designing Scalable Navigation & Taxonomy for B2B Commerce
Context
The platform required a more scalable way to manage categories, subcategories, and product relationships across a growing catalog.
The Problem
Navigation and breadcrumb structures were:
- Inconsistent
- Manually maintained
- Difficult to scale
This created friction for both users and internal teams.
My Approach
I approached this as a data + system design problem, not just UX.
The goal:
Create a system where structure is driven by product data, not manual page management
What I Did
- Defined rules-based taxonomy using structured product attributes
- Designed dynamic breadcrumb logic tied to indexed data
- Aligned Sitecore and Coveo to support automated page generation
- Reduced dependency on manual content updates
Outcome
- More consistent and intuitive navigation experience
- Reduced operational overhead for managing category structures
- Improved scalability as product catalog expanded
Why It Matters
This shifted navigation from a maintenance burden to a scalable system, enabling faster growth and better user experience.
CASE STUDY #3
Creating a Unified Chat Experience Across AI and Live Support
Context
The organization was introducing an AI-powered product chatbot alongside an existing Salesforce Live Chat experience.
The Problem
Without coordination, users would experience:
- Multiple chat entry points
- Disjointed conversations
- Repeated information requests
This created confusion and degraded user trust.
My Approach
I focused on experience continuity over feature delivery.
The goal:
A seamless transition between AI and human support
What I Did
- Defined unified entry point for chat experiences
- Designed approach to pass user context between systems
- Shifted strategy from UI-based enhancements to API-driven integration
- Partnered with engineering and vendors to align on scalable architecture
Outcome
- Clear, consistent chat experience for users
- Reduced friction when escalating from chatbot to live agent
- Established foundation for future AI-driven support enhancements
Why It Matters
AI is only valuable if it integrates cleanly into existing experiences.
This ensured AI enhanced the experience rather than fragmenting it.