Divisi
where the advice comes from
Advice you can check.
Here is exactly how it is made.
Divisi's crop recommendations are not opinions and they are not a language model's guess. They come from an implementation of the FAO's Global Agro-Ecological Zones method, run against published soil, climate and terrain data for the specific point a farmer pinned. This page explains the whole chain, including what it cannot tell you.
Every agritech product claims its advice is data-driven, and almost none of them will tell you which data, or what happens when the data is thin. If you are considering putting your programme's farmers on a platform, that is the first thing you should be able to check.
Published science, read at one point
Every recommendation starts with the coordinates of one field and reads the environment there: soil properties from global soil mapping, two decades of daily climate reanalysis for temperature and rainfall, and terrain from elevation models. These are the same public datasets used in agricultural research — the work is not in owning them, it is in reading them correctly at the scale of a single smallholding and knowing what they cannot resolve.
A suitability cascade, not a score somebody invented
The crop's potential is reduced step by step by each thing the field imposes on it — temperature through the growing cycle, water supply against demand, soil chemistry and physical constraints, terrain — in the order and manner the FAO's agro-ecological zoning framework sets out. What comes out is a suitability class and an attainable yield with the reasons attached, so a recommendation can always be interrogated: not '85 out of 100', but which factor cost what, and where.
A range, because a single number would be a lie
Climate averages carry no information about the year you are actually about to have. So alongside the expected yield, Divisi runs the same plan — same field, same crop, same planting date — through a drier and a wetter season, and reports the band. A crop with a narrow band is a crop that behaves whatever the year does, which is frequently the more valuable thing to know.
What happens when a farmer taps 'check my farm'
Read the point
Soil, climate and terrain for the pinned coordinates are assembled into one environmental picture of that field.
Run every candidate crop
Each crop in the catalogue is run through the cascade against that picture, at each plausible planting date, and ranked.
Return the reasons
The farmer gets a class, an expected yield, a range and a planting window — with what limited the crop, in words.
it doesn't work alone
One season, seen from every side
Divisi is one thing, not a bundle. Here is what this connects to — and why that matters more than any single feature.
Get Divisi free — set up in 3 minutes
Pin your farm once. From then on the advice, alerts, diary and ledger work anywhere — even with no signal in the field.
What we don't claim
The limits, plainly. Global soil and climate data resolve to a grid cell, not to a field boundary — so Divisi knows the soil of your neighbourhood better than the soil of your particular hollow, and a farmer who knows their land will sometimes be right where the model is wrong. Yields are biophysical potential at a stated input level, not a promise of what you will harvest; management, seed quality and luck all sit between the two. We publish the method and the reasoning but not the parameter tables, because those are the engine. And we have not run a randomised trial of Divisi's effect on yields — when we do, we will publish what it found, including if it is disappointing.
Questions worth asking
Keep reading
For organisations
Running an outgrower or contract-farming programme?
Ground-truth visibility across member farms, remote agronomic support, fewer field visits — on the app your farmers already use free.