E-commerce · UX redesign · Two-stage project
Redesigning a plant store's online experience — from the homepage to the filter that helped shoppers find what they came for
Plant shoppers are looking to be inspired, and then to find what they want without friction. This project tackled both problems in sequence: first, redesigning a homepage that was losing customers at the door; then, adding a shopping filter that made browsing purposeful instead of endless.
Role
Designer · Researcher · Copywriter
Industry
E-commerce / Retail
Platform
Web (desktop)
Team
UX Team of 1
Homepage Redesign
Feature design
Survey research
Card sorting
A/B testing
Usability testing
Copywriting
46
Survey respondents across both stages
6 of 7
Preferred the redesigned homepage in A/B testing
2
Research methods used to validate filter taxonomy
0
Terminology confusion after card sort refinement
THE PROBLEM
Two moments where the website was losing shoppers
The company was losing sales on two fronts. First, the homepage wasn't giving plant shoppers a reason to stay — no inspiration, no brand voice, no quality imagery. Second, shoppers who did stay found themselves endlessly scrolling through products with no way to narrow down what they were looking for. Both problems pointed to the same root cause: the website wasn't designed around how plant shoppers actually think and browse.
SHARE PROBLEM STATEMENT
Plant shoppers want to be inspired, and then find exactly what they came for — without wasting time. The website was failing at both.
Homepage redesign — getting shoppers in the door and keeping them there
RESEARCH
Who is the online plant shopper?
I distributed a survey to 46 respondents via SurveyMonkey, Reddit, and Facebook plant communities. The data pointed to a clear profile: mostly millennials — the largest and fastest-growing plant buyer demographic — who shop online at high rates and hold websites to an aesthetic standard.
Users found most plant websites uninspiring — poor navigation, bland photos, and not enough plant information to feel confident buying.
Common theme, survey responses
INFORMATION ARCHITECTURE
Fewer categories, more breathing room
Comparing the original and redesigned IA revealed the original homepage had too many options and too much noise. The redesign used fewer, broader categories — reducing cognitive load and giving each option visual space to breathe.
A/B TESTING + ITERATION
Tested with 7 users, updated after every session
Seven volunteer testers compared the original and redesigned homepage via Zoom screen share. Six of seven preferred the redesign. After each session, I updated the prototype before the next — iterating on theme colors, font, brand voice, copy, company story, and navigation simplicity.
01
WHAT CHANGED
Theme colors, typography, and brand voice updated to feel warmer and more plant-specific
03
WHAT CHANGED
Company story section added — plant shopping is a personal, values-driven experience
02
WHAT CHANGED
Navigation simplified — fewer categories, less scrolling, more space per item
04
KEY LEARNING
Shoppers want categories fast when they know what they want — but brand and story keep them coming back
Shopping filter — helping shoppers find exactly what they came for
FROM STAGE 1 TO STAGE 2
The homepage got people in. The filter helped them find what they came for.
Homepage research revealed that plant shoppers are goal-oriented once they're engaged — they want to move quickly to the right product. Without a filter, that meant endless scrolling. Stage 2 addressed that directly: add a shopping filter that lets users narrow down by the terms and categories that already made sense to them.
COMPETITOR ANALYSIS
What other plant and retail sites were doing
I analyzed competitor filter experiences across features, facets, and layout to understand conventions and identify gaps. This informed what filter options to include and how to structure them without overwhelming the user.
CARD SORTING
Letting users define the categories
I used words and phrases pulled directly from the stage 1 survey to create card sort prompts — ensuring the filter categories reflected how plant shoppers actually talk about plants, not how a retailer might categorize them. The card sort was conducted remotely in Figma.
WHY THIS MATTERED
The goal was zero terminology confusion — if shoppers had to decode a filter label, the filter had already failed.
PROTOTYPING
A filter panel built around how shoppers browse
The filter was designed as a persistent left-panel layout — visible without requiring extra clicks, and scoped to each list page category rather than applied universally. Usability testing showed that a universal filter caused confusion; a category-scoped filter felt natural and purposeful.
USABILITY TESTING
Task-based testing with multiple users
I conducted multiple rounds of usability testing, giving users a set of tasks to complete and observing where the filter helped or created friction. After each round, I asked users to share their overall experience before iterating.
→ Category order resequenced based on how users scanned the panel
→ Duplicate categories identified and merged
→ Color swatch added — users expected visual color selection, not a text list
→ Top tabs added for faster navigation between filter groups
KEY LEARNING
Scoped filters outperform universal ones
A filter applied universally across the whole site created confusion. Scoping the filter to each list page category meant users only saw relevant options — and that made the browsing experience feel intentional rather than overwhelming.
Shoppers follow a mental task list. When the filter matched that pattern, they didn't feel like they had to do a lot of clicking — they just found what they came for.
Synthesis, usability testing
FINAL DESIGNS
Homepage and filter — a complete browsing experience
The final designs across both stages work together: the homepage creates the emotional pull that brings shoppers in, and the filter gives them the efficiency to find exactly what they're looking for once they're browsing.
01
Remote testing only due to the pandemic, no budget, limited survey tools
03
WHAT STAGE 1 UNLOCKED
The homepage research surfaced the filter need — following that thread made stage 2 a natural next step, not a separate project
02
APPROACH
Persistence, empathy, and patience — found users through communities, worked within the tools available
04
KEY LEARNING
Good research doesn't just answer the current question — it tells you what to look at next
REFLECTION
