How technology is helping AgFintechs assess production performance, price farmer risk individually, and reward better farming practices
I once spent a day with Erai Maggi, the lesser-known cousin of Blairo Maggi, Brazil’s so-called King of Soybeans. At the time, Erai had already built an operation of around 800,000 acres in Mato Grosso.
But what made the day memorable wasn’t the size of the farm. It was that a small team from Rabobank was shadowing him the entire time, firing off question after question, with Erai answering each one off the top of his head.
That part isn’t unusual. Farmers, even big ones, still carry most of their operation in their heads. Ask about a particular field, harvest or machinery problem from several years ago, and many can recall the details instantly, especially if it was a painful memory.
What was unusual was that Rabobank could dedicate an entire team of analysts to spending the day on his farm. For a farmer the size of Erai Maggi, the bank spared no expense getting to know its customer. Not just for regulatory box-ticking. It made straightforward business sense.
The average farmer does not receive the same red-carpet treatment. There is no team of analysts walking the fields, no long discussion about planting decisions and no detailed assessment of how well the operation is managed.
The bank may know the crop, the region, the size of the property, the collateral and whether previous loans have been repaid. The farmer himself largely disappears.
Basket Case
Most farmers get treated like the commodities they grow. They’re priced as a basket of risk, with the weaker and less visible farmers pushing up the cost for everyone else. The best farmers end up subsidising the worst ones.
Banks have good reasons to be wary of farming. It is seasonal, exposed to weather and full of things that can go wrong between planting and payment, even when the farmer makes the right decisions.
The deeper problem is that banks have never been particularly good at measuring this risk. They can see the land, check repayment history and ask for guarantees, but struggle to see whether the farmer plants on time, uses inputs well and produces consistently.
Banks therefore fall back on what they can measure, usually land, collateral and scale. This naturally favours larger farmers, who have more assets to pledge and loans large enough to justify detailed analysis.
Two producers may grow soybeans in the same municipality but operate very differently. One may plant early, control costs and consistently produce above-average yields. The other may plant late, use inputs poorly and deliver erratic results.
Without enough information, both can still look much the same to a traditional lender, leaving the stronger farmer paying more than his individual risk profile would justify.
Ground Truth
That gap is starting to close as farming becomes more digital. Satellite imagery can show when a crop is planted, how it develops and when it is harvested. Farm-management systems, connected machinery, digital marketplaces and electronic transactions add further evidence of how the business is being run.
Together, they create a digital trail that lenders can follow across thousands of farms. One strong harvest may be luck, but consistent performance across different conditions says much more about the farmer’s ability to manage risk.
Digital technology makes production risk easier to measure and helps distinguish regional problems from those linked to the way an individual farm is managed.
By combining this operational data with traditional financial information, AgFintechs can assess not only the borrower, but also the productive operation expected to repay the loan.
Farmers already manage different fields in different ways. Finance is now starting to apply the same thinking to individual farmers.
A smaller farmer with a strong production record may be a safer borrower than a larger neighbour with more land but weaker management. Credit can begin to follow capability rather than just acreage and collateral.
Better information allows lower-risk farmers to receive larger limits, faster decisions or better rates, while weaker applicants can be monitored more closely or left out of the portfolio.
The lender gets a better-performing loan book, while the farmer receives finance that reflects how the operation is run. Good farming begins to matter more than size alone.
That can create an upward spiral. Better-financed farmers can invest in seed, inputs, irrigation, storage, machinery, technical advice and AgTech solutions.
Those investments improve productivity and resilience, while producing even more information through digital purchases, connected machinery and management systems.
The farmer becomes easier to understand, improving the next credit decision and potentially lowering the cost of future finance. Better data supports better credit, better credit supports better farming, and better farming produces more data.
The old model often works in reverse. Poor visibility raises the cost of credit, limiting the investment needed to improve both production and visibility, and reinforcing the belief that agriculture is simply too risky.
Growing Apart
Brazil has also seen what happens when agricultural distribution and credit try to scale before anyone finds a replacement for local knowledge.
The traditional input retailer often knows the farmer personally: who pays late but always pays, who is expanding too quickly and which family has managed the same property well for generations.
It is not a sophisticated credit model, but it contains valuable information that rarely appears in formal records. Companies such as AgroGalaxy and Lavoro set out to bring regional retailers together under larger platforms, gaining scale in purchasing, distribution and farmer finance.
The challenge is that consolidation can weaken the relationships on which many original credit decisions depend. The customer remains on the system, while much of the context disappears.
The problems later faced by these businesses have several causes, including leverage, working-capital pressure and difficult market conditions. But they raise a wider question: what happens when exposure to thousands of farmers grows faster than the ability to understand their production risk?
The old retailer model is personal but difficult to scale. Consolidation creates scale but risks losing the personal touch. Data-driven underwriting may be the missing bridge, preserving local visibility across a much larger network.
The same logic could also help solve one of AgTech’s biggest problems: adoption. Farmers often have to pay today for benefits that may only appear after harvest. Embedded finance can help close that gap by matching the cost more closely with the expected return.
An AgTech company with a trusted network of farmers and good visibility into production risk does not need to become a bank. It can partner with a bank, credit fund, input supplier or specialist lender.
Its data can improve underwriting, its platform can distribute the finance and the finance can help farmers adopt the technology. Farm-management platforms, satellite companies and grain marketplaces each have a different view of whether the farmer is likely to produce and repay.
Used properly, that information can make embedded finance a natural part of the business model rather than an unrelated product added later.
Beyond the Basket
AgFintech may provide the financial layer that helps the rest of AgTech scale.
For investors, Nubank offers a useful comparison. It started with a simple credit card and grew by serving people who had long been overcharged and underserved by traditional banks.
Agriculture offers a similar opportunity, with one important difference: AgFintechs can potentially understand not only the borrower, but also the productive activity expected to repay the loan.
The same farm data may eventually support other sources of revenue, from crop insurance and sustainability-linked credit to traceability, environmental service payments and the measurement and verification of farming practices.
A digital record showing that a farmer is a good credit risk may also help demonstrate compliance with evolving ESG standards, including how the farm protects soil, uses water more efficiently or reduces chemical inputs. Once the farm becomes visible, the same information can do several jobs.
Twenty years ago, knowing one farmer properly might have required a team of international analysts following him around Mato Grosso for the day.
Technology can increasingly provide a similar level of visibility across thousands of farms. Ordinary farmers no longer need to disappear inside a risk basket, and credit can begin reflecting how well they farm rather than simply how much land they own.
Banks once knew the basket. AgFintech may finally allow them to know the farmer.
Thanks for reading.
KFG
Kieran Finbar Gartlan is an Irish native with over 30 years experience living and working in Brazil. He is Managing Partner at The Yield Lab Latam, a leading venture capital firm investing in Agrifood and Climate Tech startups in Latin America.