The problem
Agricultural lending has been stuck in a 1990s paradigm — annual reviews, rigid amortisation schedules, and underwriters who have never set foot on a farm. The result: farmers borrowing at the wrong times, for the wrong amounts, at rates that don't reflect the actual risk profile of their operation.
FarmFund was built to fix this. The brief was to design a financing platform that understood farming on its own terms — not as a generic small business, but as a capital-intensive, seasonally-driven, relationship-heavy industry.
The core insight: farmers don't have cash flow problems, they have timing problems. A platform that lends against the growing season rather than the calendar year changes everything.
Approach
We started with eight weeks of farmer interviews across three states. The patterns were consistent: distrust of banks, preference for local relationships, willingness to share operational data in exchange for better terms, and deep frustration with paperwork that didn't map to how farming decisions actually get made.
From this, we defined three design principles:
Seasonally-aware — every touchpoint knows where you are in the growing year. Loan origination, payment scheduling, and dashboard defaults all shift with the season.
Land-first underwriting — soil quality, yield history, and local weather patterns feed the risk model. The application asks different questions than a bank would.
Relationship-visible — the platform surfaces your history with FarmFund over time. Farmers should see their track record, not just their balance.
The design system
FarmFund uses a warm amber accent palette — deliberate contrast to the cold blues of traditional banking. The typography pairs a utilitarian mono stack for data-heavy screens with a more expressive serif for onboarding and marketing surfaces.
The component library was built in Figma and shipped as a React package consumed by both the borrower-facing app and the internal underwriting tools — one system, two audiences.
Outcomes
Underwriting time dropped from an average of 11 days to 2.6. Default rates in the first cohort tracked below industry average by 22 basis points. And the thing we're most proud of: farmers report that the application process felt like it was written for them, not adapted from something else.