Yield Risk Forecast
Yield Risk Forecast
Building Treefera's Projects Systems
for Risk Intelligence
Building Treefera's Projects Systems
for Risk Intelligence
Overview
Overview
Treefera is an AI native, first mile intelligence platform. It turns satellite imagery, climate models, and ground data into decision grade insight for agricultural and commodity supply chains, across three pillars: Market Intelligence, Risk Intelligence, and Environmental Intelligence.
I led the design of Projects, the part of the platform where teams run and manage their own analysis. Projects had to work for very different people with very different goals, so we designed it around a template system rather than one fixed view. Within that system, we designed Yield Risk Forecast end to end, the first ready made template built on the platform, made specifically for financial services teams who need to quantify downside risk across a commodity portfolio.
This case study covers both: how Projects was designed to flex around four very different users, and the design of Yield Risk Forecast as the first template within it.
Treefera is an AI native, first mile intelligence platform. It turns satellite imagery, climate models, and ground data into decision grade insight for agricultural and commodity supply chains, across three pillars: Market Intelligence, Risk Intelligence, and Environmental Intelligence.
I led the design of Projects, the part of the platform where teams run and manage their own analysis. Projects had to work for very different people with very different goals, so we designed it around a template system rather than one fixed view. Within that system, we designed Yield Risk Forecast end to end, the first ready made template built on the platform, made specifically for financial services teams who need to quantify downside risk across a commodity portfolio.
This case study covers both: how Projects was designed to flex around four very different users, and the design of Yield Risk Forecast as the first template within it.
Role
Role
Lead Product Designer
Lead Product Designer
Team
Team
Risk Intelligence
Risk Intelligence
Timeline
Timeline
PoC: 2 Weeks
PoC: 2 Weeks
Release: 1 month
Release: 1 month
Skills
Skills
Research
Research
AI Discovery
AI Discovery
AI Ideation
AI Ideation
AI Prototyping
AI Prototyping
Testing
Testing
The Opportunity
The Opportunity
Plenty of Data, No Insights
Before this, Treefera gave people plenty of data: forecasts, risk scores, maps. But there was no dedicated space where a team could run their own analysis, save it, and come back to it over time. Every insight lived wherever it was generated, disconnected from a person's actual portfolio or ongoing work.
That gap is what Projects needed to fill. But it couldn't just be a place to view results. It had to work for very different people, doing very different jobs.
Before this, Treefera gave people plenty of data: forecasts, risk scores, maps. But there was no dedicated space where a team could run their own analysis, save it, and come back to it over time. Every insight lived wherever it was generated, disconnected from a person's actual portfolio or ongoing work.
That gap is what Projects needed to fill. But it couldn't just be a place to view results. It had to work for very different people, doing very different jobs.
The Challenge
The Challenge
4 Users, 1 System
Before designing anything, we had to understand who Projects was actually for.
Before designing anything, we had to understand who Projects was actually for.
Traders & Analyst
Weekly production forecasts before market consensus forms
Weekly production forecasts before market consensus forms
Financial Services
Quantify downside risk exposure across commodity portfolios
Quantify downside risk exposure across commodity portfolios
Supply Chain
Replace fragmented manual assessment with real-time stability signals
Replace fragmented manual assessment with real-time stability signals
Agriculture & Forestry
Plot-level evidence for carbon, deforestation, and practice verification.
Plot-level evidence for carbon, deforestation, and practice verification.
Design Problem
How do you build one space that lets each of these users get to something built for them, without forcing everyone through the same experience.
How do you build one space that lets each of these users get to something built for them, without forcing everyone through the same experience.
The Solution
The Solution
A Template System
Projects was designed around templates instead of a single fixed view. Users pick a ready made template built for their persona, configure it, and run it in minutes. If nothing fits, they build a custom one through a short guided form, pre filled with what's known about their role and portfolio.
Ready made templates cover the common needs. The custom flow catches everything else, and gives the team a steady signal for what to build next.
Projects was designed around templates instead of a single fixed view. Users pick a ready made template built for their persona, configure it, and run it in minutes. If nothing fits, they build a custom one through a short guided form, pre filled with what's known about their role and portfolio.
Ready made templates cover the common needs. The custom flow catches everything else, and gives the team a steady signal for what to build next.
The First Template
The First Template
Yield Risk Forecast
Yield Risk Forecast
Yield Risk Forecast
Yield Risk Forecast was the first ready made template the team built. It was designed specifically for financial services teams, risk managers and lenders who needed to quantify downside risk across a whole commodity portfolio, not just look at one location at a time.
The rest of this case study covers how it was built.
Yield Risk Forecast was the first ready made template the team built. It was designed specifically for financial services teams, risk managers and lenders who needed to quantify downside risk across a whole commodity portfolio, not just look at one location at a time.
The rest of this case study covers how it was built.
Background
Background
The First-Mile Visibility Gap
Organisations usually have good visibility into the middle and end of their supply chains: logistics, warehousing, distribution. But the first mile, where commodities are actually grown, stays opaque.
That's a costly gap. Disruptions at the source ripple through the whole chain, and by the time they're visible, it's often too late to react.
Organisations usually have good visibility into the middle and end of their supply chains: logistics, warehousing, distribution. But the first mile, where commodities are actually grown, stays opaque.
That's a costly gap. Disruptions at the source ripple through the whole chain, and by the time they're visible, it's often too late to react.
60%
60%
of supply chain risk originates at the first mile, shaped by weather volatility, land condition, and yield unpredictability.
Design Process
Design Process
From Ambiguity to a Working System
The challenge was surfacing deeply technical, multi-layered agricultural data in a way financial services and risk teams could act on with confidence.
The challenge was surfacing deeply technical, multi-layered agricultural data in a way financial services and risk teams could act on with confidence.
1
Stakeholder alignment & domain immersion
Worked closely with data scientists and product leads to understand the underlying models, what data could actually say and with what confidence interval.
Worked closely with data scientists and product leads to understand the underlying models, what data could actually say and with what confidence interval.
2
Competitive and analogous research
We looked financial risk dashboards, climate platforms, and geospatial tools, for interaction patterns users already trusted.
We looked financial risk dashboards, climate platforms, and geospatial tools, for interaction patterns users already trusted.
3
Information architecture
Worked closely with data scientists and product leads to understand the underlying models, what data could actually say and with what confidence interval.
Worked closely with data scientists and product leads to understand the underlying models, what data could actually say and with what confidence interval.
4
AI quick prototyping
Using AI tools to rapidly generate and iterate on layouts, we were able to test multiple directions with stakeholders and users in a much shorter time than traditional wireframing would allow.
Using AI tools to rapidly generate and iterate on layouts, we were able to test multiple directions with stakeholders and users in a much shorter time than traditional wireframing would allow.
5
Hight-fidelity design and PoC
Production-ready module with responsive states, interaction patterns, and specs for engineering handoff.
Production-ready module with responsive states, interaction patterns, and specs for engineering handoff.
What Was Shipped
What Was Shipped
Project Outlook
A plain language portfolio summary with four KPI cards: yield forecast, forecast versus baseline, downside risk probability, and area at elevated risk. This sits at the top of the page, so fast moving users get the answer before scrolling.
A plain language portfolio summary with four KPI cards: yield forecast, forecast versus baseline, downside risk probability, and area at elevated risk. This sits at the top of the page, so fast moving users get the answer before scrolling.
Risk Map
A geospatial view of yield risk across every location, clustered by risk level, colour coded, and filterable by scope and tier.
A geospatial view of yield risk across every location, clustered by risk level, colour coded, and filterable by scope and tier.
Yield Forecast Chart
Historical and projected yield against the US average, with a benchmark toggle and a confidence band.
Historical and projected yield against the US average, with a benchmark toggle and a confidence band.
Regenerative Agriculture
Portfolio level regenerative practice score distribution, plus a practice adoption breakdown covering things like crop rotation, cover cropping, and no-till.
Portfolio level regenerative practice score distribution, plus a practice adoption breakdown covering things like crop rotation, cover cropping, and no-till.
Project Locations Table
Full location level data, expected yield, yield risk rating, downside probability, sortable, filterable, and paginated.
Full location level data, expected yield, yield risk rating, downside probability, sortable, filterable, and paginated.
Key Decisions
Key Decisions
4 Decisions shaped this module:
4 Decisions shaped this module:
Outcomes and Impact
Outcomes and Impact
0 to 1 Intelligence Product
Delivered the first fully designed intelligence module on the Treefera platform, establishing a reusable pattern for Market and Environmental Intelligence pillars to follow.
Delivered the first fully designed intelligence module on the Treefera platform, establishing a reusable pattern for Market and Environmental Intelligence pillars to follow.
Design System Foundation
The components built for Yield Risk Forecast became the base design system for the entire Projects section.
The components built for Yield Risk Forecast became the base design system for the entire Projects section.
Stakeholder Alignment
The PoC resolved long-running ambiguity about how intelligence products should be structured, giving the product and engineering teams a shared, concrete reference point.
The PoC resolved long-running ambiguity about how intelligence products should be structured, giving the product and engineering teams a shared, concrete reference point.
Customer and Investor Validation
The designed module was used directly in customer discovery conversations and investor presentations, receiving positive feedback on the clarity of the intelligence surface.
Customer and Investor Validation
The designed module was used directly in customer discovery conversations and investor presentations, receiving positive feedback on the clarity of the intelligence surface.
Customer and Investor Validation
The designed module was used directly in customer discovery conversations and investor presentations, receiving positive feedback on the clarity of the intelligence surface.
Copyright © 2026 Cristina Amat
Copyright © 2026 Cristina Amat
Copyright © 2026 Cristina Amat
Copyright © 2026 Cristina Amat