I designed Leafy to give plant owners one clear answer when something goes wrong, not ten conflicting opinions, not a generic care card, but an actual diagnosis.
Leafy is a plant diagnostic platform that turns a yellowing leaf into a clear recovery plan. No guesswork. No jargon. Just a calm answer and a next step.
ROLE
Product Designer
TYPE
Self-Initiated Concept
TIMELINE
5 Weeks
PLATFORM
Web Dashboard
GENERAL INFO
About Leafy
Leafy is a houseplant diagnostic platform designed to take the anxiety out of plant care. Instead of telling users what their plant is, Leafy tells them what is wrong with it and exactly what to do next. It combines AI-powered visual diagnosis with a conversational care assistant, built around a warm, botanical design language that feels calm rather than clinical.
DESIGN PROCESS
Founder-led, decision-driven, validated early.
Founder-led process focused on clarity, restraint, and long-term behavior change.
THE PROBLEM
The gap between knowing and understanding
Plant care is not short on information. Between Reddit threads, YouTube tutorials, and care apps, there is more advice available than most people could ever read. And yet plants still die. Owners still panic. And the same yellowing leaf still sends someone to Google at midnight, sifting through contradictory answers that leave them more confused than when they started.
The problem is not a lack of information. It is a lack of clear, contextual diagnosis.
BUSINESS GOAL
Design a plant care product people trust when it matters most.
Most plant apps optimise for discovery: identifying plants, building collections, browsing species. The goal with Leafy was different. Build a product that people reach for the moment they notice something is wrong with their plant, and leave feeling like they know exactly what to do.
USER PROBLEM
Plant care feels like guesswork, not guidance.
Plant owners genuinely want to keep their plants alive, but when something goes wrong they have nowhere trustworthy to turn. Existing apps tell them what the plant is, not what is wrong with it. Generic care guides cover the average plant, not their specific one in their specific environment. Every decision feels uncertain.
COMPETITIVE ANALYSIS
Competitive Landscape and Behavioural Insights
To understand where Leafy should exist, I studied leading plant identification, care scheduling, and diagnostic apps, focusing not on features but on how their product philosophies shape user behaviour over time.
The goal was to identify what supports long-term plant care habits and what silently creates confusion, abandonment, or over-reliance on generic advice.
DESIGN PHASE
From insight to interface
With the problem framed and competitor analysis complete, the focus shifted from analysis to making. I started exploring early interface concepts for Leafy's core diagnostic experience.
Before diving into flows and systems, I explored the experience through early visual concepts, using screens to validate tone, hierarchy, and the emotional clarity of the AI interaction. The question at this stage was simple: does this feel calm, or does it feel clinical?
WIREFRAMES
Designing the structure before the surface
Before moving into visual design, I explored low-fidelity wireframes to validate structure, hierarchy, and cognitive load across every key screen.
The goal at this stage was not aesthetics or completeness. It was to ensure each screen answered a single user question and reduced the need for interpretation. A plant owner in a moment of worry should never have to figure out what to do next. The interface should already know.
These wireframes helped pressure-test the diagnostic flow and remove every unnecessary decision from the user's path.
TYPOGRAPHY AND COLOUR
Typography
Leafy's visual system was designed to feel warm, botanical, and editorially rich without ever feeling clinical or sterile. Every typographic decision supports calm readability and emotional warmth over visual noise.
High legibility on small mobile screens in any light.
Serif warmth that feels editorial, not technical.
Organic tone that avoids clinical or scientific intimidation.
Playfair Display is the brand typeface. Chosen because it carries the warmth and editorial quality of a high-end plant shop or nature magazine. When paired with Inter for all UI text, the combination reads as knowledgeable and calm rather than cold or corporate.
Color System
Leafy's colour system was designed to feel warm, alive, and instantly readable. Every colour carries semantic meaning, covering the full range from the calm of a thriving plant to the urgency of a struggling one.
AI LEVERAGE
Designing with AI as a creative partner
But based on what I documented from real creative teams, here is what changes when the tool understands the work.
AI-Assisted Research and Pattern Discovery
AI was used to accelerate early-stage research by synthesising large amounts of plant care, UX, and behavioural data, identifying what makes plant owners abandon care apps and what drives long-term engagement.
How AI helped:
Analysed plant identification, care scheduling, and diagnostic apps across the market.
Identified recurring UX failure points: identification without guidance, generic care cards, and anxiety-inducing alerts.
Compared best practices across wellness, health, and nature app categories.
Consistent illustration generating AI app
Instead of sourcing or manually creating botanical illustrations for every screen, I built a custom illustration generator using AI, trained on Leafy's warm earthy visual language to produce consistent fine-line botanical art on demand.
How AI helped:
Created a prompt-driven illustration system tuned to Leafy's botanical aesthetic.
Reduced dependency on manual illustration work across 12 screens.
Generated consistent fine-line botanical art across all screens and flows.
Maintained visual coherence with the same stroke weight, palette, and organic geometry throughout.
Rapid Prototyping with Figma Make
AI was heavily used during prototyping to visualise the full Leafy product experience early, not just individual screens but complete end-to-end diagnostic journeys before a single component was finalised.
How AI helped:
Quickly generated interactive prototypes to test the snap-to-diagnose flow end to end.
Explored multiple layout directions and care report structures in parallel.
Visualised complete user journeys early and gathered feedback before visual polish.
Design system variables were wired in from day one, so there was no visual drift across screens.
ONBOARDING
Discovering your plant's unique care blueprint
Leafy's diagnosis experience is not a simple identification tool or a generic care guide. It is a contextual diagnostic ritual designed to understand how your specific plant is responding to its environment, its light, its water, and its soil. Instead of making users learn botany, Leafy translates everything into immediate, practical care guidance.
ONBOARDING
Earn trust before asking for effort
Leafy's onboarding is designed to feel like a gentle introduction, not a questionnaire. The goal is not data collection. It is to quickly understand the plant owner's context so Leafy can reduce guesswork from day one.
