The Data Beneath the Harvest

When you sit down to eat, you rarely think about the decisions that put the food on your plate. Someone chose what to plant and when. Someone prepared the soil. Someone drove a tractor across a field at exactly the right moment in the season. Someone advised a farmer on how much fertiliser to apply and where.

Each of those decisions generated data. Across the continent, agtech platforms are capturing information about soils, machine use, weather patterns, input application, and farmer behaviour. Scientific institutions have spent decades sampling land, calibrating remote sensing imagery, and building datasets that didn’t exist before. Equipment moving across fields generates signals about where farming activity is happening and when.

And yet most of that data rarely makes it back to the farmer in a form they can use. We have built the infrastructure to collect the data but not the infrastructure to turn it into decisions.

That is the paradox at the heart of African agriculture right now. We are not short of data. We are short of the capacity to turn it into decisions that reach the people who need it most: the smallholder farmers feeding a continent.

This is a problem that iSDA and Hello Tractor have spent years working on from opposite ends of the same value chain. The two organisations bring complementary data sources and one shared conviction about what the sector needs to do next.

What Our Data Actually Looks Like

Hello Tractor is connected to thousands of tractors and combine harvesters across more than twenty countries, generating over nine billion data packets every year. That stream tells us where farming activity is happening, when in the season, what equipment is being used, and how intensively land is being worked. We use it to forecast demand, direct tractor contractors to where they are most needed, and build financial products like our Pay-As-You-Go tractor financing: a credit model that wouldn’t exist if we hadn’t first understood the patterns in how tractors move across the continent. Data, in our world, became a product that has put equipment in the hands of farmers who could never have accessed it through a bank.

At iSDA, data starts underground long before any farmer interaction. iSDAsoil is one of Africa’s most detailed soil datasets, built over years of field sampling and mapped at 30-metre resolution across the continent. It is the foundation of everything we build. But soil data alone doesn’t help a farmer. It has to be combined with information about what they are growing, what yields they are targeting, what inputs they can access, and what the season looks like. That combination is what powers Virtual Agronomist: a WhatsApp-based advisory tool that delivers personalised, science-backed recommendations to smallholder farmers throughout the growing season. In eight cohort analyses across crops and regions, farmers using VA saw an average 52% yield increase compared to traditional practice. This happened not because the soil data was sufficient on its own, but because it was translated into something actionable.

The Gap Between Data and Decision

The closer you look, the clearer it becomes that  both organisations are working on the same underlying problem from different angles.

Hello Tractor gets farmers into the field. We know when the land is prepared, and increasingly, we are beginning to experiment with advising farmers on when to plant and when to harvest, drawing on satellite data, rainfall patterns, and what farmers in similar conditions are growing. The data is there. The question is how to turn it into guidance a farmer can act on.

iSDA turns soil signals into seasonal decisions. We know what the ground can support. The question is how to reach the farmers who need that knowledge, especially those without smartphones, without consistent connectivity, and without a direct line to an agronomist.

These are not separate problems. They are the same problem, approached from opposite ends of the value chain.

What Happens When the Data Connects

The question of APIs is where this gets practical. Many organisations working with farmers already have their own platforms and their own farmer-facing applications. They don’t need another chatbot. What they need is the intelligence layer: the agronomic recommendations, the soil data, and the decision support delivered through the infrastructure they already have.

That is exactly what iSDA’s API makes possible. And it is exactly the kind of integration that Hello Tractor’s ecosystem is positioned to carry to farmers at scale, through the booking agents and tractor contractors who are already on the ground, already trusted, and already part of how farming decisions get made.

There is something that doesn’t always get named in discussions about agricultural data: the responsibility that comes with collecting it. When a farmer registers a plot, shares what they are growing, and receives advice calibrated to their specific field, they are trusting the organisations behind that system with information that matters to their livelihood. That trust has to be earned and protected.

At iSDA, farmer data is used solely to enrich the recommendation it came from, and is never shared with third parties without explicit consent. iSDAsoil, by contrast, is open and freely available to anyone, because we believe foundational scientific data should function as shared infrastructure, not proprietary advantage.

At Hello Tractor, the data our network generates informs decisions about equipment deployment, credit access, and farmer advisory, but it does so within frameworks designed to protect the individuals whose activity makes that data meaningful.

As the sector grows and data collection deepens, how organisations handle this responsibility will increasingly define who farmers and partners choose to work with.

The data beneath the harvest is not one dataset. It is soil, machines, advisory intelligence, and the pattern of a tractor booking that tells you where demand will be next season. It is the soil profile that tells you what nutrients a field needs. It is the recommendation that reaches a farmer in time to act on it.

None of this data is useful in isolation, but all of it becomes powerful when it connects.

What We’re Building Towards

Africa’s agricultural future will not be built by any single organisation. The problems are too large, the geographies too varied, the farmer populations too diverse. And the data generated across the sector is too fragmented, too siloed, and too rarely shared across the organisations that could collectively make the most of it.

What progress will be built on is the willingness of organisations working in the same ecosystem to stop treating data as a competitive asset and start treating it as shared infrastructure. That shift benefits the entire value chain. Growers make better decisions. Input providers understand where their products are going and what results they are generating. Equipment owners deploy more efficiently. Off-takers and policymakers gain visibility into what is actually happening at farm level across the regions they serve.

This is the shift the sector needs to make. The door is open for others working in this space to continue building with us.