UX / UI Case Study
A mobile app that replaces "I just water it when I remember" with a care routine that actually sticks. Built from 22 interviews with real plant owners.
TL;DR
Most people struggle to care for their plants, and existing apps fall short. I recruited 12 plant owners from the nursery I worked at, listened to what was actually getting in their way, and designed a companion app that replaces guesswork with confidence.
The Problem
People weren't neglecting their plants out of laziness. They genuinely didn't know what their plants needed, and existing apps weren't making it any clearer.
The opportunity wasn't to build another plant encyclopedia. It was to build something that told people exactly what to do, and when, in a way that felt personal rather than clinical.
Most apps sent reminders that didn't reflect a plant's actual needs — so they felt arbitrary and easy to ignore.
Care advice online was conflicting and dense, creating anxiety instead of confidence.
Owners weren't forgetful — their environment simply wasn't prompting them at the right moment, in the right way.
Most people didn't know their plant had specific requirements until something had already gone wrong.
Early Thinking
Before any pixels, I mapped out what the app could feel like in its loosest possible form — a calendar overview, checkbox-driven care, and a photo-based way to add new plants. The hand-drawn quadrants below each paired a screen idea with the behavioral problem it was trying to solve.
Research & Discovery
I had something most solo concept projects don't have: real domain context. I was still working at the local nursery where I recruited participants, which meant I wasn't pulling strangers from a panel — I was talking to real plant owners in the context where they actually bought and thought about plants. That proximity made the interviews honest, and the patterns obvious.
The Users
From the interviews, three core user types emerged — each with different needs, but all sharing the same underlying frustration: existing tools weren't meeting them where they were.
New to plant care. Just wants simple, reliable instructions to keep their plant alive without becoming a researcher.
Wants to improve skills and knowledge, but finds the sheer volume of information online overwhelming and contradictory.
Already knows the basics. Values efficiency and quick access to information over long explanations.
Design Decisions
Every core feature was chosen to remove a specific friction I'd heard named in interviews — not to build yet another plant encyclopedia.
A glanceable calendar showing which plants need care on which day — so people can restructure their schedule around trips or busy stretches, not be surprised by them.
Each plant has a simple checkbox next to it, so caring for it feels satisfying and finished — no more wondering "did I already do that?"
Upload or snap a photo of a new plant to identify it and auto-populate its care profile — removing the biggest point of anxiety for beginners.
My Plants list auto-sorts so the plants that need care soonest surface at the top — making the next action always obvious.
Hi-fidelity Product
The final interface kept a tight visual system — one primary action per screen, consistent use of the sage + fuchsia palette to signal care states, and generous white space so the content could actually breathe. Nothing here is decorative; every component earns its place by solving a behaviour I heard named in research.
Usability Testing
Two things emerged from usability testing that required real reworks, not just tweaks. Each had a clear problem that was consistent before the fix — and gone after it.
Users were confused by the wording. Most attempted to tap the photo of the plant instead of the "Yes" button at the bottom of the page.
Reworded the title to a more intuitive question, and increased the stroke weight on the confirm button for visual clarity. Tap-to-photo confusion disappeared.
When the app failed to recognize a plant, users wondered what to do next. No clear recovery path meant drop-off at a critical trust moment.
Added a Re-diagnose button that redirects the user to a new attempt page. Recovery feels deliberate instead of broken, and trust in the identification stays intact.
Brand System
The visual system balances a naturalistic sage-and-pine palette with a confident fuchsia accent — signalling "this is editorial, not childish" while keeping the primary CTAs unmissable. Playfair Display anchors the identity; Montserrat handles the work.
What It Taught Me
Working at the nursery while building this gave me something most concept projects don't have — real domain context. I wasn't just designing for a user type I'd read about. I was talking to them every day, watching them pick up plants they didn't know how to care for, and listening to the questions they actually asked.
That proximity shaped every design decision, and it's the reason the research felt grounded rather than generic. Plantopedia was a blast to work on, and it pushed my UX skills further than I expected — the kind of project that reminds me why I love this work.
Let's talk about the friction you've been quietly noticing.