Moss Robotics
Automating data collection and visualization on farms
My role
I was the lead UX designer at Moss, an agricultural tech startup specializing in robotics.
- I designed two new products from blank screens to a pair of visually beautiful apps.
- The first is an intuitive mobile app that monitors data collection in the field.
- Its desktop counterpart visualizes the collected data.
- I also created the new visual identity of Moss and set up their design system.
The two apps are currently in development.
↳ November 2023- March 2024
↳ Interaction design, prototyping, wireframing, visual design, responsive design, accessibility, user interviews
↳ 2 founders, 1 web developer
What is the value of tech on farms?
Moss is an agricultural tech startup focused on making tree grower’s lives easier. With their specialization in technology such as autonomous robots and sensor kits equipped with GPS and Lidar (Light Detection and Ranging), Moss aims to take typically labor-intensive tasks, like counting trees or spraying herbicide, happen in a fraction of the time. This helps farmers reduce labor costs, allowing workers to focus on more complex tasks around the farm. The increased accuracy of the autonomous technology helps farmers to make informed and accurate forecasts for agricultural planning.
A look in the field
Farm workers will either attach the Moss sensor kit to a tractor, or put it on a backpack. They'll take it through the field and use the mobile app to monitor data collection. Then, they will upload the collected data to the desktop app with a click of a button.
Important terms to know
How the sensor kit and control app fit into the system:
The role they play in the system:
Unique challenges:
- Time constraints: I was the sole designer, tasked with making two apps from scratch in the span of 2.5 months.
- Asks from farmers: make the interface streamlined and visual-forward.
3. The users for the mobile app have limited English/Spanish language fluency and technology fluency. The mobile app in particular had to be incredibly straightfoward and visual.
The challenge
How might we design a set of interfaces to collect, monitor, and visualize data?
My solution
A pair of apps to collect and visualize farm data
Control App (mobile)
A clear course of action
The landing page focuses on the immediate action that a user needs to take: connecting to the sensor kit, followed by selecting a mission.
Monitor progress at a glance
A streamlined view of essential information helps users quickly determine what needs to get done while in the field.
Inventory app (desktop)
View data visually
Data can appear in context on a map view. Users can adjust data layers, filter by plant ID, or zoom in to see more detailed information.
View block-level information, and toggle on/off different layers of data.
Filter for specific plants
Annotate farm maps
Farmers can upload drone maps of their farm, draw the boundaries of blocks in their field, and add information to each block.
Schedule missions
Using the annotated maps in the system, the workers in the farm office can schedule missions. These scheduled missions will be synced to the control app.
Streamlining information
The speed of the tractor with the sensor kit attached is one of the most important metrics to keep track off. High speeds can impact the quality of data collection, so users must ensure that they are within the right window (1-2 mph).
During a user interview, one of our farmers recommended that we try to emulate some of the visuals found commonly on farms. The primary users of the mobile app are farm workers, who primarily speak indigenous languages with limited Spanish and English. So, I started exploring visuals that reflected speedometers found on tractors and other farm equipment.
The evolution of the speedometer visual illustrates my process of paring down the information and emphasized the color cues for maximum readability.
Sometimes less information is better. We learned that our users out in the field would primarily care about overarching mission progress, rather than more granular information.
One of the farmers we talked to said:
“The operator is tracking many things, so [the mission in progress] area needs to be clean… the speedometer is a visual workers are used to.”
This feedback informed the final direction of this panel: one that visually emphasized the most crucial pieces of information.
Visual identity: "beautiful, and not intimidating"
Takeaways & next steps
Defining, priotizing, and designing two apps in less than 3 months was a daunting task. There were so many fascinating aspects of the problem space and explore design concepts that I would’ve loved to explore with more time.
Rigorous user testing
The biggest constraints for this project was time and resourcing (I was the only designer working on this). After discussing goals with the founders, we prioritized defining and designing the app, relying on previous research and domain knowledge from the team. With more time, I would’ve stress tested the ease-of-use for the control app with farm workers out in the field. This would also allow us to pinpoint more opportunities to incorporate visuals in the control app.
Rehaul visual identity
With more time I would have liked to spend more time refining and cleaning up the visual identity and design elements, particularly to ensure specs like color, typography, and icons are standardized and precise.
