Case Study

Rope & Cord Website Redesign

Redesigning an eCommerce navigation for clarity, confidence, and conversion.

Project Overview

Role: Lead Designer

Timeline: July 2022 - September 2022

Tools: Figma, Microsoft Clarity, Google Analytics, Canva

Focus: Navigation redesign, UX wireframes, full desktop and mobile UI, usability insights

Project Scope: Simplify the website’s navigation structure and homepage flow to make it easier for users to start shopping quickly, shop by use-case, and find the right material without needing to know the exact name.

The Problem

Rope & Cord had a deep product catalog — dozens of categories, materials, thicknesses, and types of rope and cord — but no clear entry point for new or casual users.

Users often got stuck at the very beginning of their journey:

  • Unsure where to start unless they already knew product specs

  • Navigation dropdowns were overwhelming and redundant

  • No paths for people browsing by application (e.g. “camping” or “decor”)

  • “Fishing” as a use case had poor visibility despite being one of the top-selling segments

Project Outcomes

✓ Added “Shop by Use” and “Shop by Fiber” as primary entry points

✓ Created a dedicated flow for use-specific products

✓ Improved homepage-to-product-page conversion by 22%

✓ Bounce rate dropped by 17% within the first month of launch

Research & Insights

Methods Used:

  • Click behavior tracking via Microsoft Clarity

  • Internal stakeholder interviews (customer support + product team)

  • Analytics deep-dive via Google Analytics

  • Competitor UX audits (e.g. REI, West Marine, Home Depot)

Insights:

  • New users often arrived through search but left quickly after landing on the homepage

  • Many users didn’t know the names of materials, but knew their project type

  • “Paralysis by options” was a common pattern — users scrolled, hovered, but didn’t click

Key UX Goals

✓ Reduce navigation complexity — remove duplication and streamline top-level categories

✓ Introduce “Shop by Use” paths — let users browse by intent (e.g., Fishing, Camping, Home Projects)

✓ Add “Shop by Fibers” navigation for users with technical knowledge

✓ Highlight key use cases like landscaping, marine, equine, etc. earlier in the experience

Wireframes & Ideation

Built wireframes for revised navigation system across desktop and mobile

  1. Created dropdown menu with icon-based entry points for Use and Material (fiber)

  2. Built new homepage modules: “Uses Cases,” “Fiber Types,” “Featured”

  3. Mapped flows for multiple user journeys:

    • “I’m shopping for a project”

    • “I need a specific rope type”

    • “I’m browsing for ideas”

Visuals & Prototyping

Final UI emphasized:

  • Emphasis on the search feature for users who know what they want

  • Clean layout with bold entry points into “Use” and “Material” paths

  • Modular homepage cards for specific use-cases

  • Mobile-first dropdown menus with intuitive icons and collapsible filters

  • Visual consistency across all categories to reduce cognitive load

Usability Testing

Pre-Launch Testing:

  • 4 moderated sessions via Zoom with new users

  • Tasks included:

    1. “Find rope for a towing”

    2. “Shop by material without knowing the name”

    3. “Navigate to marine ropes without using search”

Key Results:

  • 4/4 completed all tasks faster than baseline

  • Avg. time to product page reduced by 38%

  • Most common feedback:

    “I didn’t realize it could be this easy. The ‘shop by use’ just makes sense.”
    “I finally know where to go without Googling rope types.”

Post-Launch Observations:

  • Marine and Landscaping category saw a 29% increase in traffic

  • “Shop by Use” funnel showed stronger engagement on mobile

  • Clarity heatmaps confirmed deeper interaction with dropdown menus

Key Takeaways

What I Learned:

  • ✓ Navigation isn't just about structure — it’s about user psychology and intent

  • ✓ Giving users multiple ways to browse (by use or material) increases confidence

  • ✓ Too many options leads to no action — clarity drives conversion

What I’d Do Differently:

    • Push for even earlier wireframe testing with real users

    • Add on-page filtering and search refinements for even faster product discovery

    • Explore personalized entry points (e.g., “Recommended for You”)