Product Design Case Study
Designing an AI-Powered Smart Farming Platform That Helps Farmers Make Better Decisions
CropCore reimagines how smallholder and commercial farmers understand their land. By turning scattered sensor data, weather patterns, and crop imagery into clear, timely recommendations, the platform helps farmers act with confidence — spending less time interpreting dashboards and more time growing.
Role
Product Designer
Project Type
Concept / Freelance
Duration
6 Weeks
Platform
iOS & Android
Tools
Figma
FigJam
Overview
A calmer, clearer way to run a modern farm
Farming today generates more data than ever — soil sensors, satellite imagery, market prices, and forecasts. Yet most of it lives in disconnected tools that demand interpretation. CropCore consolidates these signals into a single, decision-first experience, translating complexity into a short list of things worth doing today.
The Challenge
Data everywhere, clarity nowhere
Farmers described a familiar tension: the tools meant to help them were adding cognitive load. Dashboards showed charts, not answers. Alerts arrived without context. Decisions were reactive, made under time pressure and uncertainty. The core problem wasn’t a lack of information — it was the absence of guidance.
Product Vision
From raw data to confident decisions
CropCore should feel less like an analytics platform and more like a trusted advisor. Every screen answers a single question — what should I do, and why? The vision is a system that earns trust through transparency, learns from each farm, and quietly does the heavy lifting so farmers can focus on the field.
Understanding the Landscape
Learning how farmers actually decide
Conversations with farmers and agronomists revealed decisions shaped by intuition, weather, and word of mouth as much as by numbers. Trust is hard-won and easily lost. Connectivity is inconsistent in the field. Any recommendation has to be explainable, timely, and grounded in the realities of the land — not just statistically sound.
Product Principles
The beliefs that guided every decision
Clarity over completeness
Show the one thing that matters now, not everything that could be shown.
Every recommendation is explainable
AI earns trust by showing its reasoning, never by asking for blind faith.
Respect the farmer’s time
Fewer taps, fewer decisions, faster answers — the interface stays out of the way.
Resilient by default
Works in low-connectivity fields and syncs gracefully when signal returns.
Earn trust incrementally
Start with small, verifiable wins before asking farmers to rely on automation.
Product Strategy
A single loop that compounds over time
Rather than a set of features, CropCore is built around one continuous decision loop — each pass making the next recommendation sharper.
01
Monitor Farm
02
Identify Issues
03
Receive AI Insights
04
Take Recommended Action
05
Track Progress
06
Improve Future Decisions
Design Process
Designing the End-to-End Experience
Nine connected flows, each starting from a real problem and resolved with a deliberate design decision.
01 — Onboarding
Creating a Frictionless Onboarding Experience
Problem
Farmers abandon setup when asked for too much technical detail up front
Design Decision
A progressive, map-first setup that learns the farm over time instead of all at once
02 — Command Center
Building a Personalized Farm Command Center
Problem
A generic dashboard buried the few things that actually needed attention today.
Design Decision
A prioritized home that surfaces today’s key actions and adapts to each farm’s season.
02 — Command Center
Building a Personalized Farm Command Center
Problem
A generic dashboard buried the few things that actually needed attention today.
Design Decision
A prioritized home that surfaces today’s key actions and adapts to each farm’s season.
Add UI Screen
Add UI Screen
Add UI Screen
03 — Weather
Transforming Weather into Actionable Decisions
Problem
Raw forecasts don’t tell a farmer whether to irrigate, spray, or wait.
Design Decision
Weather is paired with a clear recommendation and the best window to act.
04 — Operations
Simplifying Daily Farm Operations
Problem
Tasks, notes, and logs were scattered across paper and separate apps.
Design Decision
A single daily checklist that ties every task back to the insight that created it.
Add UI Screen
Add UI Screen
Add UI Screen
Add UI Screen
Add UI Screen
05 — Diagnosis
AI-Powered Crop Diagnosis
Problem
Identifying a disease early often requires expertise farmers can’t reach in time.
Design Decision
Point-and-scan diagnosis returns a likely cause, confidence, and next steps.
06 — Scale
Managing Farms at Scale
Problem
Multi-plot operators lost the big picture jumping between fields.
Design Decision
A portfolio view ranks every field by urgency, with drill-down on demand.
Add full-width dashboard showcase
07 — Collaboration
Enabling Collaborative Farming
Problem
Farm work is a team effort, but tools assumed a single solo user.
Design Decision
Shared farms with roles, assignments, and a synced activity log for the whole crew.
Add UI Screen
Add UI Screen
Add UI Screen
Add UI Screen
Add UI Screen
08 — Commerce
Connecting Commerce with Farming
Problem
Buying inputs meant leaving the app and losing the context of each decision.
Design Decision
Recommendations link directly to the exact inputs needed, right in context.
09 — Market
Helping Farmers Sell Smarter
Problem
Farmers often sell at the wrong moment, with little visibility into demand.
Design Decision
Price trends and demand signals suggest the best time and place to sell.
Add UI Screen
Add UI Screen
Add UI Screen
Add UI Screen
Design System
A system built for consistency at scale
Typography
A serif display face paired with a neutral sans for UI — a tight scale from 96px headlines to 16px captions.
Color Tokens
A near-black foundation with layered surfaces, hairline borders, and a single restrained green accent.
Spacing
An 8px base rhythm with generous 80–160px section spacing that lets the content breathe.
Components
Cards, chips, and controls share one radius scale and concentric corners for a cohesive feel.
Design Tokens
Every value — color, radius, elevation — is defined once as a token and reused everywhere.
Interaction Patterns
Gentle motion, soft transitions, and consistent feedback keep the experience calm and predictable.
Success Metrics
How we’d measure whether it works
Projected targets that would define success — spanning adoption, daily value, and long-term trust.
+64%
Activation
Complete farm setup in first session
4.2×
Engagement
Weekly active sessions per farmer
+38%
Productivity
Less time spent interpreting data
72%
AI Adoption
Recommendations acted on by farmers
2.1×
Marketplace
Increase in inputs bought in-app
+55%
Collaboration
Farms with more than one active member
89%
Retention
Season-over-season active farms
4.8/5
CSAT
Average satisfaction after 30 days
Reflection
What this project taught me
Designing CropCore reframed how I think about complex, data-heavy products. The hardest work wasn’t visualizing information — it was deciding what to leave out. Every chart I removed made the product feel more trustworthy, because it respected the farmer’s attention.
If I revisited it, I’d prototype the AI’s explanations even earlier. Trust is a design material, and it has to be earned interaction by interaction — not assumed because the model is accurate.
“The future of agriculture isn’t about collecting more data — it’s about helping farmers make better decisions with less effort.”