• Background
PlantSnap is the most popular plant identification app on both iOS and Android. It grew rapidly into the market leader it is today over the course of the last two years. I’ve been the main and only designer on the team. When I joined the project around August 2018 PlantSnap was pretty basic. The app was available on Android and had only recently been published for iOS. However, the design language was non-existent and most screens were stitched together haphazardly. The product faced many challenges and I’ve had to make decisions on the product’s direction & vision throughout my time on the project.
• Problem I: Onboarding
The user demographic was comprised mainly of people aged 50+. If you’ve tried teaching your granny how to use Skype you know how frustrating it can be for older people to adapt to software. PlantSnap featured several videos (~5min each), which painstakingly explain how to use the app. These had to go. The challenge here was how to introduce the functionalities of the product effectively to older people, without frustrating younger users who might find the product intuitive and the videos irritating.
Making a better Onboarding process
The process started with me taking a step back and try to see the bigger picture. I wanted to understand older users and their interaction with technology better. Some kids write their school reports exclusively on their phones. Our millennial generation might prefer web & desktop layouts. Then there are our grandparents, who feel proud if they manage to open Facebook.
At the time PlantSnap collected some data on its users, but nothing definitive that could build an argument. The best thing at my disposal were the user reviews. PlantSnap’s rating in the PlayStore at the time hovered around 3 stars. Qualitatively, users reported that they mostly find the videos confusing, and/or frustrating (~60% of reviews).
I combined that with my external research that focused on the way younger users interact with software and what they found “intuitive”.
Combining several studies, articles, as well as the data from PlantSnap, I came to the conclusion that Intuitive design is a completely different concept depending on the user’s age. It was far easier to design an interface for older users, while minimising the negative effects it may have on younger, more impatient users.
App onboardings vary from walk-through tutorials, to more subtle processes like stumble-upon hints. I decided we should have the more prominent walk-through, with a dismissal option for those that feel confident in finding their own way around.
Final solution:
The initial walkthrough blends with the UI as much as possible, allowing older users to "get a feel" for interacting with PlantSnap's interface. Skip is always present and allows others to jump straight into action. After receiving further feedback from users, we've included videos back into the app. It appeared some users still find value in them. However, we've pushed them deeper into the app, since they don't give value to the majority of our users.
I've also created several graphics to give a fresh first impression of the app. Users reported higher satisfaction with the onboarding process, when they were greeted with vibrant images.
• Problem II: UI Consistency
Even though the app only had 5 main screens, it still managed to make each one of them look completely different. There was no visual hierarchy. No color-coding. No spacing standard. No accessibility in the design. PlantSnap needed a lot of foundational design work to make sure it does not confuse users, while also creating a master design that will lead towards a scalable future for the app
Designing a complete UI system
The devs needed a rigid structure and a hierarchy of elements, with which to build.
So I began by interviewing the team on what their idea of PlantSnap’s brand was. I developed a short questionnaire to understand the core values of the company and apply them in the product. After reviewing the feedback I sketched out a few possibilities and went straight onto Figma.
Prototyping continues to be the leading way we test out ideas, even though the whole design community praises sketches like a religion. The team at PlantSnap is quite visual, and we prefer discussing prototype variations, rather than black & white sketches.
Looking back at this, it is very different from what the app is today. We rapidly tested several design variations relying on the PlayStore rating as our guide. The closer we got to five stars, the better our design language related to our userbase.
Developing a comprehensive system on my own in such a short timeframe left a few undefined situations within the app. That was okay, but for a time PlantSnap looked differently on iOS and Android. It still has variations today, but those are deliberate and due mainly to the technical limitations that come with Android (we support very old Android devices). I'm proud to say the current version of PlantSnap struck the right balance between the ideal design vision, and the real-world application.
The design language evolved with the app. As complexity increased, the need for pixel-perfect implementation followed. Figma was useful because it gave devs clear indications of the exact values I intented in my design. However, communication is always a weak point in the design-dev process. I went a step further than Figma and created breakdowns of elements, so that devs can understand and apply my vision better.
What PlantSnap is today
Today PlantSnap has an ever-growing system, which features more than 250 screens & interactions. It has been developed to be accessible, scalable, and functional on a variety of devices. It serves more than 30 million users worldwide, across 37 languages.
• Problem III: Low app usage
As with most utility apps, PlantSnap suffered from a relatively low usage. Makes sense - people don’t need a flashlight all the time, neither do they need to identify plants all the time. We can’t force people to take more snaps, but we had to create incentives for users to use the app more often.
Turning a hammer into a machine gun
As I was thinking of ways to improve the snap process and collection screen I noticed something interesting - people were creating Facebook groups, where they extended the PlantSnap experience and built a community. In those groups they spent a magnitude more time than they did in the app (the lifetime average session hovers around a minute, with rare exceptions above it). PlantSnappers, as they called themselves, shared pictures of their gardens and commented on other people’s photos. They shared memories and generally the love of plants.
We decided it was a good idea to help facilitate that community and allow them to exchange that love for plants within PlantSnap itself.
I identified several key functionalities that would’ve satisfied the community’s needs:
- Classic social feed / Nothing novel is adopted easily. Even the agile methodology relies on incremental change and anything radical immediately receives critical feedback from its users. The classical social feed was chosen as the main point of entry into the social functionalities, as it is already familiar and the learning curve will be minimal, including older users (most of them are familiar with Facebook by now).
- Direct messaging / If you look back far enough you’d remember the time when Facebook did not have DMs. Post comments and status updates where the only way users communicated. However, that is not enough - some things are better said in private and users prefer having 1 on 1 conversations with their friends. For these reasons DM was chosen as an absolutely necessary feature.
- Plant Journals / This is a novelty. The idea behind journals is to create a time-capsule. Users create a journal for a plant, a garden, or how to repot a cactus, for example. Then they create a chronologically-ordered posts within the journal that follow the progress. Other users could follow and interact with the journal. My thought process behind journals was based on the observation that the PlantSnappers community was very interested in the growth of their plants. Everyone wants to see their plants grow, flower, give fruit, etc. With journals everyone can enjoy the growth process.
Retrospective
Over the course of almost two years I can proudly say we’ve managed a lot in PlantSnap. I set a design system. The app is consistent and, despite their technical limitations, iOS & Android look pretty much alike. The small team we had on hand managed an incredible feat. A group of friends roughly 15 total managed to build a product that is used by millions worldwide on a daily basis.
However, there are many things left to be desired. We only managed to implement the classic social feed so far. Direct messaging has been put on hold by the company’s CEO. Journals and Plant care are in the pipeline, but we have no clear idea when these would come out. It is up to the project management now. Collecting data on app usage is also an important aspect of the app’s growth, which has largely been ignored, despite warnings by both the devs and myself.
Conclusively, it’s been an amazing journey. I it is my first truly global product - to think that something I’ve designed is used by more than 30 million people is wild!
