OVERVIEW
About Farmbeats
Microsoft FarmBeats is a platform that uses data and AI to enhance agricultural productivity. It collects real-time data from IoT sensors, drones, and satellites, helping farmers make informed decisions, optimize practices, and improve crop yields.
The Problem
The Solution
UNDERSTAND THE PROBLEM
User Research
To understand farmers' workflows and identify essential app features, we employed three user research methods to uncover their challenges, pain points, and goals.
Research Methods:
Reviewed 20+ articles to analyze challenges and advancements in agriculture.
Conducted 6 thirty minutes interviews to identify user experience constraints.
Surveyed farmers to uncover pain points and validate prior research insights.
General Insights
Mistimed operations like irrigation reduce yields by up to 50%.
Diseases and pests destroy up to 40% of crops annually, causing economic losses.
Users are highly concerns about the impact on crop diseases and pests.
Real-time nutrient monitoring boosts crop yield by 15%.
Pain Points
Based on our analysis of user research data, we have identified three major pain points in the current agricultural sector:
PAIN POINT 1:
Insufficient Monitoring of Growth Rates:
Farmers often struggle to accurately assess the growth speed and stages of crops, which is crucial for determining optimal times for fertilization and irrigation. Lack of real-time and precise growth data can lead to resource wastage and suboptimal yields.
PAIN POINT 2:
Difficulty in Early Detection of Diseases and Pests:
Farmers typically fail to detect early signs of pest infestations and diseases until significant damage is apparent. This delay in detection makes it difficult to apply control measures in a timely manner, potentially leading to severe crop losses.
PAIN POINT 3:
Inefficient Task Management:
Farmers often struggle with managing and prioritizing the numerous tasks required to maintain crop health and productivity. The lack of an integrated system to track, schedule, and monitor tasks leads to inefficiencies and missed opportunities, ultimately impacting overall farm productivity.
KNOW USER
Personas
From our user research, we identified three key user groups: small-scale farmers needing real-time data for efficient crop management, agronomists requiring precise tools for expert analysis, and product managers focused on developing user-centered agricultural solutions. These personas guided the design to meet the specific needs and challenges of each group.
Competitive Analysis
To better understand the product landscape and identify market opportunities, we conducted an analysis of 5 competing products. We assessed key features, including those we plan to implement, against common industry standards.
Our analysis revealed gaps in the market, specifically the absence of features like crop growth rate tracking, task management, and AI assistance, which we aim to emphasize in our product.
Design Principles
Based on user research insights, personas, and competitive analysis, we identified key features required by our target users and uncovered market opportunities, leading to the establishment of 4 strategic design principles to effectively meet user needs.
Simple and Intuitive Design
Highlight key information, simplify workflows, and use progressive guidance to help users start with basics and gradually master advanced features, while maintaining a tech-savvy feel.
Environmentally Friendly Design
Optimize resource management and incorporate scalability to ensure the design is sustainable and future-proof.
Seamless AI Integration
Designed an intuitive dashboard with embedded AI for real-time recommendations to reduce cognitive load, with a floating assistant readily accessible for on-demand help.
Efficient Modular Functionality
Separate core features into independent modules (e.g., task management, map views), support one-click operations and automation to improve efficiency and reduce the learning curve.
THE FRAMEWORK
User Flow
To better understand how users interact with the product, I broke down the user flow by individual features instead of mapping the entire app at once. This approach helped clarify the logic behind user actions and how each page interacts within specific features.
START DESIGN
Lo-fi Wireframe
Building on the information architecture and user flow, we created a lo-fi wireframe to outline the basic design layout and structure of the interface.
Hi-fi Prototype
After ensuring that all user needs and key features were addressed in our design, we refined the hi-fi prototype, focusing on the design details and interactions between pages. This preparation was essential for the upcoming usability testing.
THE REFINEMENT
Usability Testing
Conducted 6 remote usability tests with small-scale farmers, agronomy students, and farm-related product managers to gather feedback and refine the design.
We used severity ratings to prioritize user frustrations and friction points, leading to these 4 must-have iterations.
Design System
To support the final design, I built a comprehensive design system, establishing foundational tokens like colors, typography, and grid based on the hi-fi prototype. As I refined the design, I developed components and guidelines to ensure consistency, enhance user experience, and streamline future iterations.
FEATURES
Monitor Crop Growth
The homepage crop card now displays the crop growth rate directly.
On the crop detail page, additional growth-related information is provided, along with a linear growth chart to help users better monitor plant development.
AI Assistant
Powered by GPT-4, the AI assistant is seamlessly integrated across the app, offering targeted suggestions for crops, sensors, and tasks. Users can interact through text, photos, and voice, allowing for more accurate crop analysis based on different crop condition.
Additionally, a floating button enables easy access to the AI assistant from any page, enhancing the overall user experience.
Detect Pests
Implemented pest detection notifications to alert users immediately. Enhanced the map view to clearly display the location of detected pests, accompanied by AI-driven pest control and prevention suggestions to help users address the issue efficiently.
Task Management
The task management feature helps users make better decisions and improve crop productivity. Additionally, the AI assistant provides plant-specific suggestions that can be directly added to tasks.
Map View
Designed dedicated map views for both crops and sensors, allowing users to easily visualize their locations within the fields.
REFLECTION
Takeaways
Less is more: During the final design iteration, I realized that mobile interfaces require a focused approach. Unlike desktop designs, highlighting too many elements on a mobile screen can lead to a cluttered experience. Moving forward, I’ll emphasize key information by displaying less and subtly de-emphasizing peripheral details.
Modular Flow Design: Instead of creating a single, complex user flow for the entire app, I designed separate flows for each critical feature. This approach made it easier to clarify user logic and accelerate the design process.
Validated Pain Points: I utilized three research methods to identify key pain points, which were later validated through usability testing. Pain points may evolve, but the crucial takeaway is the importance of continuous testing to ensure the design meets the real needs of farmers and agronomists.
