Client
Duration
An AI-powered mobile app that helps people identify houseplants, understand their condition, and build a personalized care routine through explainable recommendations, reminders, and ongoing monitoring.


About the project
Houseplant owners often notice a problem but cannot understand what caused it or what to do next. Advice is scattered across different sources, while most plant apps focus on one-time identification instead of continuous care.
The Plantix is an AI-powered plant care assistant designed for both beginners and experienced plant owners. It connects identification, diagnosis, personalized recommendations, reminders, and condition monitoring within one continuous experience.
The central element of the product is the plant profile. It stores care history, current status, scheduled tasks, recommendations, and sensor data, helping users understand not only what is happening to a plant, but why it is happening and what action should come next.

Research & product decisions
The main challenge was to make AI recommendations useful and trustworthy without overwhelming beginners or limiting the control expected by experienced plant owners.
The research included a survey with 30 plant owners, interviews, an open card-sorting session with 12 participants, and corridor tests with 8 target users.
The findings showed that most people care for plants reactively: they start looking for help only after visible symptoms appear. Users also struggle to identify plants, receive contradictory advice, forget recurring tasks, and hesitate to trust AI conclusions that are not explained.
These insights shaped the product. Photo identification became the main entry point, while AI results were supplemented with confidence levels, possible causes, and prioritized actions. Monitoring data was translated into clear “normal”, “attention”, and “critical” states. Detailed history, charts, and sensor settings remained available through progressive disclosure for more experienced users.




Summary
The result is a connected product concept that supports the user from the first photo to regular, informed plant care.
The project resulted in a complete interactive mobile prototype, information architecture, key user flows, wireframes, final UI screens, and a scalable design system.
Usability testing confirmed the value of fast identification, explainable AI recommendations, and contextual reminders. Testing also revealed problems with terminology, overloaded screens, and sensor setup. The interface was refined through simpler microcopy, stronger visual hierarchy, progressive disclosure, and step-by-step configuration.
No launch or business metrics are presented because the product has not been released. The next stage would be to test long-term personalization, integrate real IoT sensors, improve AI confidence explanations, and prepare the concept for MVP development.
DISCOVER MORE





