Jeffrey Oh

Terra: a home gardening assistant

A mobile app that makes plant care simple, whatever your skill level.

Product
End-to-end mobile app
Role
Product and UX design
Duration
2 weeks
Terra app screens on a pale green background around the Terra logo: tasks, calendar, plant gallery, plant details and sign-in.

At a glance

Problem
It’s easy to forget plant care, and the leading apps either overwhelm beginners with information or lock key features behind a paywall.
Approach
Competitive analysis of 3 apps, personas by skill level, a card sort with 7 participants, task flows, sketches, a prototype and user interviews.
Outcome
The card sort split location and light into separate steps. Interviewees found plant registration smooth and liked the camera recognition; the task details page needs more work.

Background

Terra is a mobile application created with users’ time in mind. It focuses on simplicity and an elegant design, with a streamlined caretaking process to help you look after your home plants and gardens.

Problem

I created this application to simplify the logging and caretaking side of home gardening. Everybody has a lot on their minds nowadays, and it becomes easy to forget to take care of our plants. This drove me to create an application that would solve this problem in a simple and elegant way.

Goals

  1. Improve the daily diligence of home gardeners
  2. Create a simple solution for the widest range of gardening skill
  3. Make something enjoyable

UI highlights

Home screen with a sunflower weather banner for Philadelphia, a 63-day streak, and today's and tomorrow's watering tasks with round check buttons.
Calendar view for March with the 18th selected and that day's tasks listed below.
Your Plants screen grouping plant photos by room, such as Office and Patio, with an add button.
Google Lens camera view identifying an aloe vera plant, with Retake and Yes buttons.

Research

Competitive analysis, personas and a card sort.

Competitive analysis

From Seed to Spoon

Helpful features

  • Planting dates with a projected harvest date
  • Location can be specified, though not in a visually clear way
  • Archive plants that were removed or died
  • Notes per location and per plant
  • Health benefits, growing season, frost tolerance, sprout-to-harvest time, sun, water and fertilizer guidance

Takeaways

  • Visual consistency improves UX
  • Less is more

Planter

Helpful features

  • Virtual gardens for square foot gardening
  • Task schedule in a timeline format
  • Gardening guides
  • Add your own images

Takeaways

  • Lean away from information overload
  • Registering a plant should not overwhelm the user

Planta

Helpful features

  • Simple and clean UI
  • Organize and view plants by site or by plant
  • Straightforward multi-step plant registration
  • Home page is a task page, with the weather at the top

Takeaways

  • Don’t show features that require premium
  • Essential features should not be premium
  • Do not over-advertise

Selection process

When looking for competitors, I looked at the highest performers: the largest active user base with the best ratings. These 3 stuck out not only for those reasons but also for the creative utilities they provide. For example, Planter had done something I had never seen in gardening apps: a virtual garden to manage, which was their main selling point. Although I didn’t use this feature, I could still learn from their design patterns and processes.

A great amount of inspiration came from Planta, which had one of the most elegant designs. One thing I couldn’t get behind was that Planta is a freemium application, with many of the convenient features locked behind a paywall. It ruined the experience and created a divide between users.

From Seed to Spoon, while not the cleanest looking application, covered all of its bases when it came to plant care: temperature, watering methods, bugs and pests, tutorials. You name it, they had it.

Competitor app screenshot: a garden list showing an aloe vera plant with planting and harvest dates, under a Garden Plus upgrade banner.
Competitor app screenshot: a dark plant catalog grouped into alliums and cole crops, with an Add Plant button.
Competitor app screenshot: a skill level picker from Hopeless to Master, each with a plant photo.

Personas

Beginner

Can keep low-maintenance plants alive for the most part. Does not stick to a schedule and believes watering cycles are the same for all plants. Does not track plant progress at all.

Intermediate

Understands water cycles and light levels, and keeps a number of plants alive, but needs a way to organize tasks. Occasionally takes photos to track progress.

Advanced

Understands the ins and outs of gardening. Probably doesn’t need tips, but wants a way to track plant growth virtually. Keeps a journal manually or in other apps.

Card sort

Card sort results matrix. Rows are terms such as Home, Kitchen, Low Light, Mist and Aloe Vera; columns are Add a New Plant, How to Take Care, and Where Is My Plant, shaded by how many participants chose each.

I ran a card sort with 7 participants, who sorted each card into 3 categories. The primary takeaway was that words about light and place were grouped together, and words about treatment and light landed in several categories.

Using this, I put location and light on separate pages so there was less confusion around the topics.

Define

The foundations of the app, through maps and sketches.

Site map

Terra site map: login leads to the home page, which branches to calendar, profile, add task and plants pages, with plant registration steps for location, light settings, plant details and Google Lens.

User flows

Three task flows in black page boxes and yellow action boxes: register a plant, add a task to the schedule, and complete a task.
Primary use cases. Black boxes are pages; yellow boxes are actions.

I knew that covering a larger breadth of use cases would make the application feel worse overall, so I took a few use cases and worked through them carefully and in depth. These task flows gave me a reference so I wouldn’t go astray in my design process. Succinct flows let me design much faster, and there was never a point where I was wondering what was next. (At least in the early stages.)

Revised add-a-task flow: from the new task screen, users either enter task details themselves or use recommended details, then confirm.
Modified task scheduling flow. I created this flow to account for experience levels; the previous version did not cover all of my users.

Sketches

Pencil sketch of the plants page and the light selection page with options from bright to dark.
Pencil sketch of the login screen and the home task list with a weather banner.

In early ideation I was already close to my current designs. One idea I tried was a light gauge. It didn’t work out: it felt gimmicky and didn’t fit the aesthetic I was going for, which broke the coherency and consistency of the app.

Button placement and larger containers changed, but sizing and functionality stayed the same. Looking at my login page and home page, I remember how many internal debates I had about placement, current design trends, and the best combination. I couldn’t decide so early on, so I picked what felt simple and elegant. From there, all that was left was testing and iterating.

Screens

Home screen with weather banner, daily streak and the task timeline.
Where is your plant screen with photo tiles for office, patio, hall, kitchen, bedroom, bathroom and a custom option.
How bright is it screen with photo options for bright, part sun, shade and dark.
Plant details form with an image recognition button, name, notes and an optional photo.
Task details screen with a recommended task setup button, a watering time and repeat settings.
Profile screen with account details, notification toggles and statistics such as a 63-day longest streak.

Home page

This is the page I spent the most time, and tears, on. In the earliest iterations I couldn’t get the weather banner to look the way I wanted without compromising functionality and visual consistency. I ended up with a masked circular banner to ease the boxed-in feeling of square containers.

For the task timeline, the completion bubbles started out as checkboxes. I wanted the app to feel free and lighthearted while staying simple and elegant, so I softened as many elements as I could. Completed tasks are greyed out to say, “I just finished this task, it’s time to move on.”

Task details page

This page had the most considerations of any: so much research and so many iterations to create something that looks like an alarm setup. It is my biggest regret in the project and the page I’d most like to improve.

I’m proud of the auto task setup feature, but the rest of the fields lack flexibility and customization. Quick fixes would be a separate notes field, a dynamic calendar instead of the alarm-style setup, and more information for the user.

Testing

Prototype

The prototype includes pages in various states, interactions and site connections. Open the Terra file in Figma.

Interview findings

Needs

  • More information on the plant details screen
  • Optional tips and tricks for each plant
  • A more refined task creation flow

Wants

  • Name preset locations during plant registration
  • Clarity on what calendar sync does
  • A preview of the push notification

Likes and dislikes

  • Liked: plant registration is smooth and simple
  • Liked: Google Lens camera recognition
  • Disliked: daily streak design is inconsistent
  • Disliked: the calendar could be cumbersome

Insights

  • Gardening is much deeper than it seems, and may need deeper note-taking for plants.
  • Understanding what users value drives functionality and new features.
  • Balancing hidden versus visible features is tricky and needs more testing.

Final remarks: a letter to you

This case study had me stumped for a while. It was my biggest creative block, since it was a topic I wasn’t knowledgeable about and the time frame was limited.

Over the 2 weeks, about 40% of the time went into deciding what should and shouldn’t be in the design. I wanted a simple, refined experience for a problem that was in no way simple or refined, which made the task much more challenging than I expected.

The main lesson I learned was never to assume anything about time blocks, schedules and deadlines. Colleagues and mentors told me this is normal. People run into blockers, creative or otherwise. The schedule is not law and not something to overstress about. We take our losses and learn from them, so we can improve at what we do best: designing.

As for the future: visually, this project doesn’t need to go much further. I would spend time on micro-interactions and make sure visual feedback is strong enough for a good experience.

If you made it to the end of this case study, thank you for taking the time. If you’re interested in working with me, get in touch.