Cassava is the crop Ghanaian agronomy treats as needing none. It is planted where the maize was, fed what the maize left over, and lifted when the seller needs cash rather than when the buyer needs starch. Each of those is a decision, and each is being made by default. This pillar is not a growing guide; it is the four field decisions that are worth real money and are currently made backwards.
Cassava does not want its nutrients in equal parts. It wants potash first, nitrogen second and phosphate a distant third, because potassium is the nutrient that most limits its yield and, critically, the one that builds the starch an industrial buyer pays for (International Atomic Energy Agency, 2018). The compound fertiliser most available to a Ghanaian farmer is NPK 15:15:15, which is equal-share by construction, fifteen parts of each in every bag. So the bag over-delivers the nutrient cassava wants least and under-delivers the one that raises starch. Fixing the shape, potash-led, lifts both the tonnage and the dry matter, and the gain a wrongly-shaped bag leaves on the table lands in someone else’s factory as a lower starch yield.
The second half of the fertiliser problem is the rotation. Cassava is routinely planted behind a fertilised cereal on the hope of a residual effect, receives nothing of its own, and quietly mines the field’s potassium and organic matter, a decline invisible on any single season but real over five (Adjei et al., 2023). Fertilise the cassava in its own right against its expected yield, and write a soil test into any offtake contract so the buyer and the lender can see the land is not being run down.
| Nutrient | What cassava wants | NPK 15:15:15 delivers | The mismatch |
|---|---|---|---|
| Potassium (potash) | The most, and it builds starch | An equal third | Under-delivered; the yield and dry-matter loss |
| Nitrogen | The middle share | An equal third | Roughly matched |
| Phosphate | The least, by a long way | An equal third | Over-delivered; money spent on the least-wanted nutrient |
Sources: International Atomic Energy Agency (2018) on the potash-first nutrient order; Adjei et al. (2023) on rotation-driven depletion.
Weeds come first. Cassava is a poor competitor in its first three months because it grows slowly and closes canopy late, so weed control in that window is what protects the yield. Weeding already eats 50 to 80 per cent of the labour budget on a cassava farm, most of it done by hand and most of it by women.1 The lever is better-timed weeding, backed by spacing and herbicide that close the canopy sooner and starve the weeds of light.
The harvest date is a business decision disguised as a farming one. Starch content rises fast and then plateaus at around the tenth month, while the tonnage of root keeps climbing for months more. A farmer paid by weight has every reason to wait; a processor paying for dry matter gains almost nothing after the plateau and may lose quality. No spot contract mentions a harvest month, and it should: contract the month as well as the tonnage, and pay for dry matter, because the same lifting date that sets the weight also starts the two-to-three-day perishability clock that governs everything downstream (Pillar 3).
Cassava matures in 9 to 14 months, so the harvest window falls well after planting. The north has one rainy season and a concentrated planting and harvest; the south has two, which spread supply across the year. Timing the harvest to the buyer’s calendar, rather than to the seller’s need for cash, is what turns this pattern into a reliable supply.
The value ladder runs from fresh root through dried chips, gari and high-quality flour to native and food-grade or modified starch. Value, shelf life and safety all rise with processing, which is why the dry matter this pillar grows is only realised downstream (Pillar 6), and why a harvest paid by dry matter, not weight, is the field’s first claim on that value. The ordering is indicative of the progression, not a price scale.
Soil comes second. Cassava will crop on tired land where other crops fail, which is why farmers plant it last in the rotation and rarely feed it. It is still a heavy feeder, and potassium is often the nutrient that most limits its yield. A continuous cassava-maize rotation significantly depletes soil organic matter, nitrogen, phosphorus and exchangeable cations including potassium, so yields fall over successive cycles unless the soil is fed back.2 The lever is simple to say and hard to fund: replace potassium and organic matter, and stop treating the cassava plot as the place where soil goes to be mined.
Practice and cost come third. Smallholders already use climate-smart methods, with mulching, mixed farming, crop diversification and rotation all above 80 per cent uptake in surveyed cassava communities. The barriers are money and know-how: weak technical assistance, pests and diseases, and thin knowledge of the practices, each named by about 90 per cent of farmers, on top of the high cost of fertiliser.3 Mechanization stays thin, with most farm operations still done by hand and land preparation both the most-mechanized task and the one in shortest supply.4 The 2026 plan to deploy several thousand machines through shared-access Farmer Service Centres could ease that bottleneck for growers who cannot own a tractor.5
Ghana’s food-processing industry by scale: about 40 per cent small-scale and informal, where most cassava flour and gari are made, 30 per cent small and medium enterprises, and 30 per cent large-scale and multinational operators that dominate revenue and export. The field decisions this pillar sets, fertiliser shape, weeding and harvest timing, feed a chain whose value still sits mostly in small, informal hands.
NPK 15:15:15 is equal-share, but cassava wants potash first; the farmer overpays for the nutrient the crop wants least and shorts the one that raises starch, and the loss lands in the buyer's factory as low dry matter.
Every farmer aiming at an industrial buyer, and every processor whose throughput depends on the dry matter the field delivers.
Fix it at the buyer rather than the farmer: specify and finance a potash-led blend in the offtake and recover it against measured dry matter at the gate; soil-test first.
Cassava loses most to weeds in its first three months, and weeding is the biggest labour cost on the farm, so a late first weeding quietly caps the harvest.
Smallholders relying on hired labour, whose margin can turn negative when the early weed-control window is missed.
Weed early in the critical window; use recommended spacing and herbicide to close canopy faster and cut repeat weedings.
The fertiliser mistake is one of shape rather than amount: cassava wants potash first, but the common NPK 15:15:15 bag is equal-share, so it shorts the nutrient that builds starch.
Fix fertiliser at the buyer: specify a potash-led blend in the offtake and recover it against measured dry matter at the gate.
The first three months decide the crop: the early weeding window is where yield is won or lost, and it is the dominant labour cost.
Harvest to the buyer's calendar rather than the seller's: starch plateaus around month ten while tonnage keeps climbing, so contract the month and pay for dry matter.
Cassava crops tired land, but continuous cropping mines potassium and organic matter; feed the soil back or the yield drifts down.
Where this connects.The variety and clean planting material that set the ceiling are Pillar 1’s; the perishability clock that the harvest date starts is Pillar 3’s; the cost of the labour and inputs this pillar decides is Pillar 5’s; and the processing that finally pays for the dry matter grown here is developed in the value chain, Pillar 6.