GreenTrack is being shaped not only by code, but by conversations in the field. These stories document the people, observations and experiences that are helping define what the technology should become.
Stories from the people behind your food.
Photo / Field Documentation — George Ngugi. Jonathan Chomba, coffee farmer, Subukia.
A field conversation with a Kenyan Arabica coffee farmer revealed just how much information travels with a coffee harvest — from the farm, through a wet mill and a dry mill, to an auction floor, and finally to a buyer.
Read the Story →
Erah Chomba grows spinach and cabbages in Subukia. Watching her work through GreenTrack for the first time — and hearing what she asked for next — reshaped how this app thinks about usability and market access.
George Ngugi, on one of the same field visitsSoftware development paired with direct field research — meet the person behind both, and the plots he's visited to get here.
GreenTrack started as a broader idea — connecting the journey from growing food to understanding, tracking, and eventually consuming it, across farmers, chefs, diners and shoppers.
Rather than design entirely from behind a computer, field visits around Subukia, Kenya put the product in front of the people it was meant to serve.
Conversations with growers like Erah Chomba and Jonathan Chomba surfaced gaps no amount of desk research would have found — from interface friction to the need for market access.
What began as a multi-role concept narrowed into a sharper focus on the grower's side of the journey — crop tracking, plot management, crop health, and nutrition.
Features like the nutrition calculator and pest/crop-health scanning were tested directly with growers in the field, not just imagined at a desk.
Field research isn't a one-time step. Each conversation — including the ones documented on this page — continues to shape what GreenTrack builds next.
Observations from the visits documented on this page — not fabricated statistics, just what came out of the conversations themselves.
A feature is only useful if the intended user can comfortably reach and use it — something Erah's first minutes with the app made hard to ignore.
Tracking crops and harvests is valuable, but farmers also need practical support around what happens after harvesting — starting with finding a market.
Erah's request for help finding markets showed that production and selling can't always be treated as separate problems.
The nutrition calculator gave growers a way to see their harvest as food with value — not just a quantity logged in an app.
Jonathan's coffee plot exposed a problem that would have been easy to miss designing only from a developer's perspective: not every crop behaves the same way.
Not every conversation became a full article — but every photograph here is real documentation from the same GreenTrack field research trip around Subukia. Click any photo for a closer look, or see the full field gallery →
Erah Chomba, grower — Subukia
Jonathan Chomba, coffee farmer
A visit to the Sub-County horticulture office
Grace Wanjeri, at her stall
A grower, among his maize
Signing consent for documentation
Coffee rows, Subukia
Spinach, early growth
John, beneath his banana tree
Working the spinach beds
Logging a batch, in the field
Jonathan, working the rows
John Maina, among his maize
John, beneath his banana tree
George combines software development with direct field research to understand real-world problems before building technology to address them. Rather than design GreenTrack purely from behind a screen, he visited growers around Subukia — Erah Chomba, Jonathan Chomba, and others documented on this page — to see how the app actually holds up in someone's hands, on someone's plot. He doesn't only build the interface. He goes into the field to understand the person using it.
Every story here traces back to a real plot, a real grower, and a real harvest.
Jonathan Chomba, coffee farmer
Erah Chomba, grower — Subukia
John Maina, farmer — Subukia
Coffee cherries, Subukia
The app, in the field
Spinach, early growth
John's banana tree, Subukia
Coffee rows, SubukiaThe stories in this Journal come from real interviews conducted by George Ngugi around Subukia, Kenya, with the interviewee's knowledge and, where documented, their written consent. We don't invent people, quotes, or statistics — where a detail isn't confirmed, we leave it out rather than fill the gap.
That's a stricter bar than most "customer stories" pages hold themselves to, and it means we publish less often than we'd like. We think it's the only way a page like this is worth reading at all.
Erah Chomba, growerSpinach, cabbages, and a usability lesson that reshaped how GreenTrack thinks about market access.
Read the story →
Jonathan Chomba, coffee farmerFrom the farm to the mill and eventually the market, Jonathan's coffee journey shows why every harvest carries a story.
Read the story →
John Maina, farmerMaize, ginger and banana on one plot — and the same market-price problem waiting at the end of each harvest.
Read the story →
Grace Wanjeri, produce traderA stall in Subukia, and the same question every trader eventually asks: what's coming, and what state will it be in?
Read the story →
Agnes Kariuki, grocery shop ownerA grocery shop owner's question wasn't about supply. It was about what had actually been sprayed on the produce.
Read the story →
Eric Githinji grows strawberries and spinach in Subukia — a new field profile, published as documentation is still being gathered.
Read the story →
The full field archiveA produce trader, a horticulture officer, and every other frame from the same trip — the complete, filterable photo archive.
Browse the gallery →George Ngugi conducts each interview in person, with the interviewee's knowledge and, where documented, their written consent. Every quote you see attributed as a direct quote is genuinely their words; where a specific quote wasn't recorded, the story says so and paraphrases instead.