The website recruits people who have already decided to grow avocados. It says almost nothing to the person still deciding — or to the household with mature trees dropping fruit on the ground. This page sets out three tools that close that gap, and demonstrates the engine behind them.
The value proposition to a committed grower is complete — aggregation, grading, transparent pricing, shares, a member portal. That only converts someone who has already made the hard decision.
Lands on four invitations to become a member, and no answer to the questions that come first: can my land grow this, what will it cost, and when do I see money back? The funnel assumes the decision is already made, so every undecided visitor leaves.
Households and farms across Lusaka have mature trees planted years ago for shade or the pot, with fruit rotting each season. They don't identify as "avocado farmers", so nothing on the site is addressed to them — though they are the fastest members to onboard, with fruit ready this season.
No price range per grade, no yield expectation per tree, no spacing, no pollination guidance, no seasonality calendar. The tools below need this data anyway — so producing it does double duty: it fills the content gap and powers the tools.
One model — trees × age × variety × price — behind two front doors. "What are my trees worth" is the discovery door; "planning to plant" is the same engine projecting forward, including the J-curve that prospective growers consistently underestimate.
Revenue is the easy half. These are the outflows that make the J-curve dip. Every figure is yours to change — they start blank on purpose, because Ontech does not know what a seedling costs you.
Prices move. A projection built on today's price is a hope, not a plan. Press this to halve the price you entered and watch what happens to your break-even — the number that survives a bad year is the number to trust.
| Year | Age | Yield (kg) | Revenue | Net | Cumulative |
|---|
Estimate only, computed from the figures you entered. Not a forecast, not advice, and not a guarantee of income from the cooperative.
The land check and the drainage pit test depend on no cooperative data, so they are built, live, and public — under a name a farmer would actually search for.
Ten years of real rainfall, temperature, wind and soil moisture for any point in Zambia, from ERA5-Land reanalysis. Rules areas out; never rules a plot in — the grid is ~8 km, which we measured rather than assumed.
A real state machine with a real clock that walks a grower through a proper percolation test and reads the result. It is the only thing on the site that measures the grower's actual plot — so it outranks everything else.
Works from what a grower already knows: tree count, rough age, variety. No new data infrastructure. It reaches Gap 2 — the households already sitting on income — who are the fastest members to onboard because their fruit is ready this season. Variety ID starts as a static photo guide (Hass vs Fuerte vs seedling); photo upload comes later.
Same engine, projecting forward, showing the J-curve honestly: money out for three to four years before the crossover. Showing the wait builds trust and filters out the growers who would quit in year two. Ships when the price schedule exists.
Most work (map UI plus geospatial lookups), weakest claim, and the one that carries reputational risk if overstated. Decide it last.